Vous êtes sur la page 1sur 141

Trends in U.S.

Oil and Natural Gas


Upstream Costs

March 2016

Independent Statistics & Analysis U.S. Department of Energy


www.eia.gov Washington, DC 20585
This report was prepared by the U.S. Energy Information Administration (EIA), the statistical and
analytical agency within the U.S. Department of Energy. By law, EIAs data, analyses, and forecasts are
independent of approval by any other officer or employee of the United States Government. The views
in this report therefore should not be construed as representing those of the Department of Energy or
other federal agencies.

U.S. Energy Information Administration | Trends in U.S. Oil and Natural Gas Upstream Costs i
March2016

Contents
Summary..................................................................................................................................................1
Onshorecosts..........................................................................................................................................2
Offshorecosts..........................................................................................................................................5
Approach..................................................................................................................................................6
AppendixIHSOilandGasUpstreamCostStudy(CommissionbyEIA).................................................7

I. Introduction........IHS3
II. SummaryofResultsandConclusionsOnshoreBasins/Plays.....IHS7
III. DeepWaterGulfofMexico..........IHS23
IV. MethodologyandTechnicalApproach....IHS29
V. BakkenPlayLevelResults........IHS35
VI. EagleFordPlayLevelResults........IHS53
VII. MarcellusPlayLevelResults....IHS69
VIII. PermianPlayLevelResults...IHS85
IX. DeepwaterGulfofMexicoIHS103

Figures
Figure1.RegionalshaledevelopmenthasdrivenincreasesinU.S.crudeoilandnaturalgasproduction.2
Figure2PercentagebreakdownofcostsharesforU.S.onshoreoilandnaturalgasdrillingand
completion....................................................................................................................................................3
Figure3.Averagewelldrillingandcompletioncostsforthe5onshoreplaysstudiedfollowsimilar
trajectories....................................................................................................................................................4
Figure4.Costperverticaldepthandhorizontallength...............................................................................5

U.S.EnergyInformationAdministration|TrendsinU.S.OilandNaturalGasUpstreamCosts ii
March 2016

Summary
The profitability of oil and natural gas development activity depends on both the prices realized by
producers and the cost and productivity of newly developed wells. Prices, costs, and new well
productivity have all experienced significant changes over the past decade. Price developments are
readily observable in markets for oil and natural gas, while trends in well productivity are tracked by
many sources, including EIAs Drilling Productivity Report which focuses on well productivity in key
shale gas and tight oil plays.

Regarding well development costs, there is a general understanding that they are sensitive to increased
efficiency in drilling and completion, which tends to lower costs, shifts towards longer wells with more
complex completions, which tends to increase them, and prices for oil and natural gas, which affect
markets for drilling and completion services through their effect on drilling activity. However, overall
trends in well development costs are generally less transparent than price and productivity trends.
Given the role of present and future cost trends to determining future trajectories of U.S. oil and natural
gas production under a range of possible future price scenarios, it is clearly important to develop a
deeper understanding of cost drivers and trends.

To increase the availability of such cost information, the U.S. Energy Information Administration (EIA)
commissioned IHS Global Inc. (IHS) to perform a study of upstream drilling and production costs. The IHS
report assesses capital and operating costs associated with drilling, completing, and operating wells and
facilities. The report focuses on five onshore regions, including the Bakken, Eagle Ford, and Marcellus
plays, two plays (Midland and Delaware) within the Permian basin1, as well as the offshore federal Gulf
of Mexico (GOM). The period studied runs from 2006 through 2015, with forecasts to 2018.

Among the reports key findings are that average well drilling and completion costs in five onshore
areas evaluated in 2015 were between 25% and 30% below their level in 2012, when costs per well
were at their highest point over the past decade.

Based on expectations of continuing oversupply of global oil in 2016, the IHS report foresees a
continued downward trajectory in costs as drilling activity declines. For example, the IHS report expects
rig rates to fall by 5% to 10% in 2016 with increases of 5% in 2017 and 2018. The IHS report also expects
additional efficiencies in drilling rates, lateral lengths, proppant use, multi-well pads, and number of
stages that will further drive down costs measured in terms of dollars per barrel of oil-equivalent
($/boe) by 7% to 22% over this period.

EIA is already using the observations developed in the IHS report as a guide to potential changes in near-
term costs as exploration and production companies deal with a challenging price environment.

1The Bakken is primarily located in North Dakota, while the Marcellus is primarily located in Pennsylvania. The Eagle Ford and
the two Permian plays (Midland and Delaware) are located in Texas.

U.S. Energy Information Administration | Trends in U.S. Oil and Natural Gas Upstream Costs 1
March 2016

Onshore costs
Costs in domestic shale gas and tight oil plays were a key focus of EIAs interest given that development
of those resources drove the major surge in crude oil and natural gas production in the United States
over the past decade, as shown in Figure 1. The IHS report documents the upstream costs associated
with this growth, including increases associated with the demand for higher drilling activity during
expansion and decreases during the recent contraction of drilling activity.

Figure 1. Regional shale development has driven increases in U.S. crude oil and natural gas production
Crude oil production Marketed natural gas production
million barrels per day billion cubic feet per day
10 90
Rest of U.S. Rest of U.S.
80
Federal Gulf of Mexico
8 Rest of U.S shale
Permian region 70
Eagle Ford region Marcellus region
60
Bakken region
6
50

40
4
30

2 20

10

- -

Source: U.S. Energy Information Administration Drilling Productivity Report regions, Petroleum Supply Monthly, Natural Gas
Monthly
Note: Shale gas estimates are derived from state administrative data collected by DrillingInfo Inc. and represent the U.S. Energy
Information Administrations shale gas estimates, but are not survey data.

The IHS report considers the costs of onshore oil and natural gas wells using the following cost
categories: land acquisition; capitalized drilling, completion, and facilities costs; lease operating
expenses; and gathering processing and transport costs. Total capital costs per well in the onshore
regions considered in the study from $4.9 million to $8.3 million, including average completion costs
that generally fell in the range of $ 2.9 million to $ 5.6 million per well. However, there is considerable
cost variability between individual wells.

Figure 2 focuses on five key cost categories that together account for more than three quarters of the
total costs for drilling and completing typical U.S. onshore wells.2 Rig and drilling fluids costs make up
15% of total costs, and include expenses incurred in overall drilling activity, driven by larger market
conditions and the time required to drill the total well depth. Casing and cement costs total 11% of total

2Typical U.S. onshore wells are multi-stage, hydraulically fractured, and drilled horizontally. The costs identified relate, in part,
to the application of those technologies.

U.S. Energy Information Administration | Trends in U.S. Oil and Natural Gas Upstream Costs 2
March 2016

costs, and relate to casing design required by local well conditions and the cost of materials. Frac
Pumps, Equipment costs make up 24% of total costs, including the costs of equipment and horsepower
required for the specific treatment. Proppant costs make up an average of 14% of total costs and
include the amount and rates for the particular type of material introduced as proppant in the well.
Completion fluids, flow back costs make up 12% of total costs, and include sourcing and disposal of the
water and other materials used in hydraulic fracturing and other treatments that are dependent on
geology and play location as well as available sources.

Figure 2 Percentage breakdown of cost shares for U.S. onshore oil and natural gas drilling and
completion

Rig and drilling fluid


15%
23%
Casing and cement

Frac Pumps, Equipment


11%
Proppant

12% Completion fluids, flow back

Other
24%
14%

Source: IHS Oil and Gas Upstream Cost Study commissioned by EIA

Over time, these costs have changed. For example, drilling and completion cost indices shown in Figure
3 during the period when drilling and drilling services industries were ramping up capacity from 2006 to
2012 demonstrate the effect of rapid growth in drilling activity. Since then, reduced activity as well as
improved drilling efficiency and tools used have reduced overall well costs. Changes in cost rates and
well parameters have affected plays differently in 2015, with recent savings ranging from 7% to 22%
relative to 2014 costs.

U.S. Energy Information Administration | Trends in U.S. Oil and Natural Gas Upstream Costs 3
March 2016

Figure 3. Average well drilling and completion costs for the 5 onshore plays studied follow similar
trajectories
Cost by year for 2014 well parameters
$ million per well

$12
$10
$8
$6
$4
$2
$-
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

Eagle Ford Bakken Marcellus Midland Delaware


Note: Midland and Delaware are two plays within the Permian basin, located in Texas and New Mexico
Source: IHS Oil and Gas Upstream Cost Study commissioned by EIA

The onshore oil and natural gas industry continues to evolve, developing best practices and improving
well designs. This evolution resulted in reduced drilling and completion times, lower total well costs, and
increased well performance. Drilling technology improvements include longer laterals, improved geo-
steering, increased drilling rates, minimal casing and liner, multi-pad drilling, and improved efficiency in
surface operations. Completion technology improvements include increased proppant volumes, number
and position of fracturing stages, shift to hybrid fluid systems, faster fracturing operations, less premium
proppant, and optimization of spacing and stacking. Although well costs are trending higher, collectively,
these improvements have lowered the unit cost of production in $/boe.

The cost variations across the studied areas arise primarily from differences in geology, well depth, and
water disposal options. For example, Bakken wells are the most costly because of long well lengths and
use of higher-cost manufactured and resin coated proppants. In contrast, Marcellus wells are the least
costly because the wells are shallower and use less expensive natural sand proppant. Figure 4 shows, by
region, how costs for well vertical and horizontal depths have dropped over time, driving some of the
efficiency improvements characteristic of U.S. domestic production over the past decade.

The Bakken play has consistently had the lowest average drilling and completion costs of the basins and
plays reviewed in the IHS report. Improvement in drilling rig efficiency and completion crew capacity
helped drive down drilling costs per total depth and completion costs per lateral foot, since 2012.
Recent declines are partly a result of an oversupply of rigs and service providers. Standardization of
drilling and completion techniques will continue to push costs down.

U.S. Energy Information Administration | Trends in U.S. Oil and Natural Gas Upstream Costs 4
March 2016

Figure 4. Cost per vertical depth and horizontal length


Drilling Cost per Total Depth Completion Cost per Lateral Foot
$ per foot $ per foot

$250 $1,000
$200 $800
$150 $600
$100 $400
$50 $200
$- $-
2010 2012 2014 2016 2018 2010 2012 2014 2016 2018
Eagle Ford Bakken Marcellus Eagle Ford Bakken Marcellus
Midland Delaware Midland Delaware

Note: Midland and Delaware are two plays within the Permian basin, located in Texas and New Mexico
Source: IHS Oil and Gas Upstream Cost Study commissioned by EIA

Offshore costs
There are fewer than 100 deepwater wells in the Gulf of Mexico. Unlike onshore shale and tight wells
that tend to be similar in the same play or basin, each offshore project has a unique design and cost
profile. Deepwater development generally occurs in the form of expensive, high-risk, long-duration
projects that are less sensitive to short-term fluctuations in oil prices than onshore development of
shale gas and tight oil resources. Nevertheless, recent low commodity prices do appear to have reduced
some Gulf of Mexico offshore drilling.

Key cost drivers for offshore drilling include water depth, well depth, reservoir pressure and
temperature, field size, and distance from shore. Drilling itself is a much larger share of total well costs
in offshore development than in onshore development, where tangible and intangible drilling costs
typically represent only about 30% to 40% of total well costs.

According to the IHS reports modeling of current deepwater Gulf of Mexico projects, full cycle
economics result in breakeven prices that are typically higher than $60/b. Low oil prices force
companies to control costs, increase efficiencies, and access improved technologies to improve the
economics in the larger plays. Efforts are underway to renegotiate contract rates and leverage existing
production infrastructure to develop resources with subsea tiebacks. Consequently, the IHS report
forecasts a 15% reduction in deepwater costs in 2015, with a 3% per annum cost growth from 2016 to
2020. The large cost reduction in 2015 is most notable in rig rates because of overbuilding.

U.S. Energy Information Administration | Trends in U.S. Oil and Natural Gas Upstream Costs 5
March 2016

Approach
The IHS report includes the following analyses and results:

Assessment of current costs and major cost components


Identification of key cost drivers and their effects on ranges of costs
Review of historical cost trends and evolution of key cost drivers as well designs and drilling
programs evolved
Analysis of these data to assess likely future trends, particularly for key cost drivers, especially in
light of recent commodity price decreases and related cost reductions
Data and analyses to determine the correlations between activities related to drilling and
completion and total well cost

U.S. Energy Information Administration | Trends in U.S. Oil and Natural Gas Upstream Costs 6
March 2016

Appendix - IHS Oil and Gas Upstream Cost Study (Commission by EIA)
The text and data tables from the IHS Oil and Gas Upstream Cost Study are attached.

U.S. Energy Information Administration | Trends in U.S. Oil and Natural Gas Upstream Costs 7
EIA UPSTREAM COST STUDY

FINAL REPORT

Oil and Gas Upstream


Cost Study
DT007965, CO Task Assignment
Definitization Letter FY2015 #4

Prepared For:

Energy Information Administration


(EIA)
October 8, 2015

Submitted by:

IHS Global Inc.


5333 Westheimer Drive

Houston, Texas 77056

1
EIA UPSTREAM COST STUDY

Table of Contents

IHS Points of Contact:


Richard F. Fullenbaum
Vice-President Economic Consulting
IHS Economics and Country Risk
1150 Connecticut Ave NW, Suite 401
Washington DC 20036
Tel 1-202-481-9212
Email: Richard.Fullenbaum@ihs.com

Curtis Smith
Director Upstream Consulting
IHS Global, Inc.
5333 Westheimer Rd
Houston, TX 77056
Tel _1 713-369-0209
Email: Curtis.Smith@ihs.com

Project Team Members

Richard Fullenbaum (project executive)


Curtis Smith (project manager)
Min Rao
Jing Xiao
Stephen Adams
Russ Fontaine

2
EIA UPSTREAM COST STUDY

I. Introduction
The Energy Information Administration, (EIA) commissioned IHS Global Inc. (IHS) to perform a study of
upstream costs associated with key basins and plays located in the United State, namely the Bakken,
Eagle Ford, Marcellus, Permian Basin and deep water Gulf of Mexico (GOM). As explained by EIA, one of
the primary purposes of this study is to help EIA analysts with cost analyses and projections that the
organization is required to provide. Consequently, the study focused on the most active areas, and the
results include the following:

Determining current costs and major cost components,


Identifying key cost drivers and their impact on range of cost,
Reviewing historical cost trends and evolution of the key cost drivers as well designs and drilling
programs have evolved,
Analyzing the relevant data to determine future trends, particularly for key cost drivers in light
of recent commodity price decreases and related cost reductions, and
Providing data and analyses to determine the correlations between activities related to drilling
and completion and total well cost

IHS based this study on 2014 costs. However, the collapse of oil prices in late 2014 has forced reduction
of many upstream costs, thus modifying the cost structure. Consequently, this report addresses future
cost indices, including cost reductions for 2015, and how key cost drivers will continue to play a role in
changing costs.

This report begins with a discussion of summary results for the selected onshore basins and deep water
Gulf of Mexico, and then addresses methodologies and assumptions. The body of the report is
comprised of detailed discussions of costs for each basin, including the deep water Gulf of Mexico. A
large data set is also available in conjunction with this report, which includes many additional graphs
and charts not included herein. These are listed in the Appendix.

A. Background to the Study


As a result of low oil prices, US onshore oil field development had nearly come to a standstill by the year
2000. However, relatively stronger natural gas prices encouraged the drilling of vertical wells in
conventional natural gas plays and some development of coalbed methane. The shale boom began with
the Barnett Shale taking off in 2004, employing modern unconventional drilling and completion
techniques such as horizontal drilling and complex hydraulic fracturing (fracking). These techniques
evolved as they spread to other plays such as the Haynesville in Northern Louisiana, the Fayetteville in
Northern Arkansas and the Marcellus Shale in Pennsylvania and West Virginia. Increasing natural gas
prices from 2001 through 2008 also fueled this evolution.

While natural gas prices collapsed in 2008, oil prices, which had begun an upward trajectory early in the
decade, dropped as well. However, unlike natural gas, oil prices quickly rebounded, driving operators to
explore new opportunities in search of oil plays and liquid-rich gas plays containing associated
condensate and natural gas liquids (NGLs). New plays such as the Eagle Ford and Bakken became
profitable by drilling and fracking horizontal wells, tapping into the shale source rocks of earlier
productive plays.

3
EIA UPSTREAM COST STUDY

At the same time, deep water and deep formation areas offshore, once prohibitively expensive to
explore or develop had new technology and strong oil prices to encourage more difficult operations.
Moving into deeper water was accompanied by technical and commercial challenges, as was drilling into
deep formations with high temperature and high pressure (HTHP). However, with large deep water
discoveries, such as Jack in 2004, deep water exploration and development in the Gulf of Mexico
spurred ahead.

Since the advent of unconventional plays, drilling and completion of wells have continued to evolve with
their associated costs increasing commensurately. For example, short lateral lengths of just 1000 to
2000 feet have increased substantially to as much as 10,000 feet in some plays. Proppant use and
intensity of hydraulic fracturing have also increased, resulting in huge increases in well performance.
This evolution has led to significantly higher well cost (on average greater than 6 million dollars
(MM$)/well). However the associated productivity gains have offset these costs, resulting in lower unit
costs per barrel of oil equivalent (Boe) and providing better returns on investment. Operators continue
seeking the optimal return through two means: 1) persistently driving down actual costs by increasing
efficiency, and 2) trying to optimize unit costs ($/Boe) by finding the right balance between high-cost
completion design and enhanced performance.

In 2011, as commodity prices stabilized, we saw a large uptick in drilling, resulting in shortages of supply
and increased costs. To combat this trend, some operators became more vertically integrated into field
services and supplies. For example, some companies purchased or developed sand mines, water
treatment facilities, gas processing plants, pipeline infrastructure, or even drilling rigs to have primary
access to services that could ensure lower costs.

By 2014, as plays became delineated and the better performing areas were identified, the Bakken, Eagle
Ford, Permian Basin and Marcellus plays emerged as the most significant contributors to the
unconventional oil and natural gas supply and capital expenditure within the U.S. The oil price collapse
of 2014 forced changes upon the market, including capital cost reductions, downsized budgets and more
focused concentration on better prospects within these plays. Some offshore capital costs (such as rig
rates) were also being reduced, but unlike unconventional plays where capital expenditures can be
turned on and off relatively quickly, offshore development and budgeting is a longer term proposition.
Therefore, we may not see substantial changes in offshore activity levels unless low prices persist for
several years.

This study focuses on areas of intense current and forecasted activity that will have a material effect on
future production and capital expenditure; these include four onshore plays or basins, namely the
Bakken, Eagle Ford, Marcellus and Permian Basin, as well as the deep water Gulf of Mexico. No attempt
is being made to provide an apples-to-apples comparison between the onshore and offshore basins, as
the mode of capital operating expenditure is vastly different. Since this comparison is not practical,
onshore and offshore basins are discussed separately throughout the report.

B. Scope and Approach


Upstream costs analyzed within this study include capital and operating costs associated with drilling,
completing and operating wells and facilities. Some pipeline costs are included in the offshore analysis.
The analysis uses cost modeling that incorporates the following taxonomy.

4
EIA UPSTREAM COST STUDY

Onshore

1. Drilling Within onshore basins drilling comprises about 30-40% of total well costs. These costs
are comprised of activities associated with utilizing a rig to drill the well to total depth and
include:
a. Tangible Costs such as well casing and liner, which have to be capitalized and
depreciated over time, and
b. Intangible Costs, which can be expensed and include drill bits, rig hire fees, logging and
other services, cement, mud and drilling fluids, and fuel costs.
2. Completion Within onshore basins completion comprises 55-70% of total well costs. These
costs include well perforations, fracking, water supply and disposal. Typically this work is
performed using specialized frack crews and a workover rig or coiled tubing and include:
a. Tangible Costs such as liners, tubing, Christmas trees and packers, and
b. Intangible Costs include frack-proppants of various types and grades, frack fluids which
may contain chemicals and gels along with large amounts of water, fees pertaining to
use of several large frack pumping units and frack crews, perforating crews and
equipment and water disposal.
3. Facilities Within onshore basins facilities construction comprises 7-8% of total well cost. These
costs include:
a. Road construction and site preparation,
b. Surface equipment, such as storage tanks, separators, dehydrators and hook up to
gathering systems, and
c. Artificial lift installations.
4. Operation These comprise primarily the lease operating expenses. Costs can be highly
variable, depending on product, location, well size and well productivity. Typically these costs
include:
a. Fixed lease costs including artificial lift, well maintenance and minor workover activities.
These accrue over time, but are generally reported on a $/boe basis.
b. Variable operating costs to deliver oil and natural gas products to a purchase point or
pricing hub. Because the facilities for these services are owned by third party
midstream companies, the upstream producer generally pays a fee based on the volume
of oil or natural gas. These costs are measured by $/Mcf or MMbtu or $/bbl and include
gathering, processing, transport, and gas compression.

Offshore Deepwater

The rig and related costs account for 90-95% of total well costs, for both drilling and completion and
primarily includes the day rate of utilizing drilling ship or a semi-submersible drilling rig for drilling,
completing the well, and all other rig related costs, such as drilling crew, fuel, consumables, support
vessels, helicopters, logging, cementing, shore base supplies, etc.

5
EIA UPSTREAM COST STUDY

1. Drilling The drilling activity itself comprises about 60% of total offshore well D&C cost. These
costs are comprised of activities associated with drilling the well to total depth and include rig
hire fees, drilling bits, logging, casings, liners and other services, cement, mud and drilling fluids,
fuel costs, offshore support services and other services.
2. Completion Within offshore basins, this comprises less than 40% of total well costs. These
costs are comprised of completion rig hiring, well perforations and testing, completion fluid,
transportation/logistics, well stimulation and sand control, and well head equipment.
3. Injection Wells For a typical field, additional wells are drilled to reinject produced water
and/or gas in order to maintain reservoir pressure.
4. Facilities Production facilities are another major expense and may include one or more of the
following:
a. Floating facilities, such as tension leg platforms (TLP), Spars or Semisubmersible
platforms. These facilities may include topsides, production equipment, such as
compressors, separators and processing units, and capabilities to drill additional wells.
b. Sub-sea tiebacks to production facilities with customized sea floor assembly and riser
connecting platforms.
5. Operation Operation costs are primarily comprised of the lease operating expenses, which can
be highly variable depending on product mix, water depth, distance from the shore, facility size
and configuration. These costs are typically accrued and estimated on a monthly basis. Costs
include:
a. Variable operating costs, which may consist of costs associated with delivery of oil and
gas products to a purchase point or pricing hub when products leave the operator-built
pipeline and enter a transportation system controlled by a third party. Since the
upstream producer pays a fee based on the volume of oil or gas, costs are measured by
$/Mcf or MMbtu or $/bbl.
6. Transport - For new field development, a pipeline will be required to tie into existing
infrastructure from the production facilities. Such capital expenditures are borne by the
producer.

Cost Modeling

In the cost modeling, a rate was applied to determine the total cost of an item by determining a well
or facility configuration and the amount of material or labor required for each major item. The cost
for each item was summed up to obtain
the total well or facility cost.

All costs and calculations are based on


incorporating the inflation rate and are
determined using nominal dollars. We
believe that eliminating inflation
provides a better method for
determining costs going forward,
especially for offshore facilities where
construction and implementation can Figure1-1: Historical and forecasted inflation
take many years. While no adjustments

6
EIA UPSTREAM COST STUDY

to costs were made for inflation, we have included historical and forecasted inflation rates in the
event the reader desires to back-calculate costs by removing inflation (see Figure 1-1).

II. Summary Results and Conclusions Onshore Basins/Plays


A. Basic Well Design and Cost for 2014
Total well capital costs within the four onshore basin/plays (plays) are grouped by drilling, completion
and facilities (see Figure 2-1) and range from $4.9 MM to $8.3 MM. An additional $1.0 MM to $3.5 MM
in lease operating expenses may be incurred over a 20-year well life cycle and a similar amount may be
incurred for gathering, processing and transport (GPT) costs over the life of the well. Play location, well
dimension, completion, (hydraulic fracture) intensity and design determine the ultimate cost per well.
Well type (oil/gas), location, performance or amount of production and longevity determine total
operating expense.

Drilling costs include rig rental, tubulars


such as casing and liner, drilling fluids,
diesel fuel and cement. Total well cost can
vary greatly from play to play and within a
play depending on such factors as depth
and well design. Average horizontal well
drilling costs range from $1.8 MM to $2.6
MM and account for 27% to 38% of a wells
total cost. Before the expansion of
horizontal drilling within unconventional
plays, drilling costs ranged from 60% to as
Figure 2-1: Allocation of drilling and completion much as 80% of a wells cost.

Completion costs include completion liner and tubing, wellhead equipment, source water, water
additives, sand proppant, completion and perforating crews and pumping equipment rentals. Average
completion costs generally fall in the range of $2.9 MM to $5.6 MM per well, but some are higher, thus
making up 60% to 71% of a wells total cost. Completion costs in North America have risen sharply over
the last decade due to horizontal drilling, in particular due to lateral lengths becoming longer and
completions becoming larger and more complex each year.

Oil and natural gas field facilities costs include separators, flow lines, evaporation pits, batteries, roads
and pumps or compressors to push product to gathering lines. They generally fall in the range of several
hundred thousand dollars and make up just 2% to 8% of a wells costs. Often several wells are drilled
consecutively on a drilling unit or pad where each well benefits from economies-of-scale as more wells
share the same facilities. Alternatively, wells may be drilled one to a pad as operators try to hold
acreage by production by drilling as few wells as possible.

Operating expenses Due to variability, operating costs are addressed for each play. A general
discussion pertaining to the three major operating cost categories is addressed below:

7
EIA UPSTREAM COST STUDY

Lease operating expense: These costs are incurred over the life of a well and are highly variable
within and between the plays. Oil plays, for example, have particular activities such as artificial
lift that make up a large portion of the cost, whereas natural gas prone plays do not. Lease
operating expenses range between $2.00 per Boe to $14.50 per Boe, including water disposal
costs. Wells with more production will generate more cost over the life of the well. Deeper
wells in oil plays will generate more costs than shallower ones.
Gathering, processing and transport: These costs are associated with bringing each mcf of gas
or barrel of oil to a sales point. Fees are governed mostly by individual contracts that producers
enter into with third-party midstream providers and can be highly variable. Typically, operators
with larger positions within a play are able to negotiate better rates. Each product has its own
set of requirements and associated costs:
O Dry gas, which requires no processing, incurs the lowest costs at approximately
$0.35/Mcf for gathering and transport to a regional sales point with a differential to
Henry Hub price ranging from $0.02 to $1.40 per mcf.
O Wet gas includes NGLs that require fees for processing, fractionation and
transport. Associated gas within the oil plays is generally classified as wet gas and
requires processing as well. Gathering and processing fees typically range from $0.65 to
$1.30 per Mcf. Fractionation fees range from $2.00 to $4.00 per bbl of NGL
recovered. NGL transportation rates range from $2.20 to $9.78 per bbl.
O Oil and condensate can be transported through gathering lines at a cost ranging between
$0.25 and $1.50 per bbl. Trucking is much more expensive with costs ranging between
$2.00 and $3.50 per bbl. Operators will also need to transport oil longer distances to
refineries, either by pipeline or by rail which creates a price differential to the play
ranging from $2.20 to $13.00 per bbl.
Water disposal: Most of the flow-back water disposal expenses from fracking operations are
included in capital costs. After 30-45 days (when most of the flow back water has been
removed) these expenses would then be classified as operational and would include residual
flow-back water and formation water. Specific expenses are related to the water-oil or water-
gas ratios and disposal methods and include reinjecting water into water disposal wells, trucking
and recycling programs. Thus, costs are highly variable, ranging from $1.00 to $8.00 per bbl of
water.
In addition General and Administrative costs (G&A) are included as operating expenses and can
add an additional $1.00 - $4.00 per boe.

Land acquisition There are typically four ways that operators are able to acquire an acreage position in
one of these plays, and each may greatly affect the overall cost of operation:

Aggressive entrant The Operator acquires a large land position (usually over 100,000 acres)
within a play based on initial geologic assessments before the play begins to develop and long
before the play is de-risked or pilot programs begin. Although operators are able to acquire
land quite cheaply ($200 -$400 per acre), those who follow this strategy often acquire land in
speculative plays that never become economic, and hence incur a substantial risk that the
development of the acreage will never come to fruition.

8
EIA UPSTREAM COST STUDY

Legacy owner Because the plays discussed herein generally occur in mature basins with
historic conventional production, operators basically inherit an acreage position in the play by
virtue of already being a participant in conventional production. While this may save substantial
costs, these operators may not have necessarily landed in the sweet spots or better areas of the
play.
Fast follower Operators who do not have the capacity to lease land may choose to form a Joint
Venture (JV) with a company who has an acreage position. This will typically occur after the play
has been de-risked and appears to be viable. However, at this stage sweet spots may not be
completely delineated and operators could end up with a sub-standard position. Costs will
typically be 10 to 20 times higher in this scenario than for initial entry. Depending on the
number of acres required per well, this could add additional costs of $1 - $2 MM per well.
Late Entrant Typically late entrants will be motivated to enter a play once the sweet spot has
been delineated and the play completely de-risked. They will pay a premium of 3 to 4 times that
of the fast follower which will include potential drilling locations as well as producing wells. In
order to meet economic thresholds, these operators will be looking for tight down spacing,
stacked laterals and other upside potential.

While acquiring land in any of these plays can add substantial upstream costs, each operator pursues
the strategy that they believe will provide the best financial returns. For purposes of this study we will
address this issue in each play by providing historical transaction costs and an estimated well spacing to
determine the added cost that theoretically could be added to the cost of each well for an operator
entering a play during a specific year. We should bear in mind, however, that once the money has been
spent to acquire a land position, the acquiring operator will treat these as sunk costs and therefore
when performing go-forward economics these costs will not be included.

B. Geological and Technical Considerations by Play


The close relationship of average horizontal
well depth (including both the vertical and
horizontal portions) and the respective drilling
costs for each play is portrayed in Figure 2-2.
While the amount of fluid and proppant in
each play greatly influences the overall
completion costs, the correlation of proppant
and fluid volumes to completion cost is not as
strong (see Figure 2-3). Other factors such as
pressure, use of artificial proppants and frack
stage spacing also influence completion costs.
Figure 2-2: Depth and drilling cost by play
Since its inception, the Bakken play has been
known for long wells and big completions. The average true vertical depth (TVD) of 10,000 feet is fairly
constant throughout the play where drilling costs average $2.4MM, but is slightly deeper in frontier
areas where drilling costs are $2.6MM. Although the Bakken was the first play to move to long lateral
lengths of approximately 10,000 feet with as many as 30- 40 frack stages, the use of proppant and fluid
per foot is much lower than other plays. While average proppant use is lower than other plays, costs

9
EIA UPSTREAM COST STUDY

are comparable, as the Bakken uses more of


the higher-cost artificial and resin coated
proppants, which drive the completion costs
from $4.4 MM to $4.8MM. Moderate to high
pressure gradients also drive completion
costs higher and require the use of a higher
artificial proppant mix.

Unlike the Bakken, true vertical depths in the


Eagle Ford vary greatly from 6,000 feet in
shallow oil-prone areas to more than 11,000
Figure 2-3: Proppant and completion cost by play feet in the gas-prone areas. Lateral lengths
are fairly constant, averaging 6000 feet.
Overall, drilling costs range from $2.1 MM to $2.5 MM. Like the Bakken, proppant costs per pound are
higher due to heavy reliance on artificial proppant. Completion costs range from $4.3 MM in the more
oily areas to $5.1MM in gas prone areas. Overall, pressure is high in this play, but more so in the deeper
gas prone areas, which also drive completion costs and artificial proppant use up.

Wells in the Marcellus are shallower, averaging 5,000 to 8,000 feet in depth with a lower formation
pressure gradient. Lateral length is highly variable, ranging from 2,500 to 7,000 feet. While operators
would prefer to drill the longer laterals, smaller leases and drilling units dont always allow this to
happen. Drilling costs are fairly uniform ranging from $1.9 MM to $2.1 MM. Proppant costs in the
Marcellus are low as less-expensive natural proppant is popular. However proppant amounts are higher
in the Marcellus than in other plays and are highly variable, resulting in completion costs ranging from
$2.9MM to $5.6MM.

The Permian Basin contains two primary sub-basins, the Midland Basin and Delaware Basin, many
diverse plays and complicated geology of stacked formations in desert conditions. Most unconventional
wells are horizontal with expensive completions, similar to the Eagle Ford (averaging $6.6 MM to
$7.6MM), but may be small vertical wells accessing the stacked pay zones in the Sprayberry costing only
$2.5 MM per well. Formation depths vary
from 7,000 to 10,000 feet. Lateral lengths
and frack designs differ largely by region
and play with completion costs ranging
from $3.8 MM to $5.2 MM. High proppant
use is the norm.

C. Key Cost Drivers


Overall, 77% of a typical modern
unconventional wells total cost is
comprised of just five key cost categories
(see Figure 2-4):

Figure 2-4: Primary cost

10
EIA UPSTREAM COST STUDY

Drilling Related Costs: (1) rig related costs (rig rates and drilling fluids), and (2) casing and
cement.
Completion Related Costs: (3) hydraulic fracture pump units and equipment (horsepower), (4)
completion fluids and flow back disposal, and (5) proppants.

(1) Rig related costs are dependent on drilling efficiency, well depths, rig day rates, mud use and diesel
fuel rates. Rig day rates and diesel costs are related to larger market conditions and overall drilling
activity rather than well design. Rig related costs can range from $0.9 MM to $1.3 MM making up 12%
to 19% of a wells total cost.

(2) Casing costs are driven by the casing markets, often related to steel prices, the dimensions of the
well, and by the formations or pressures that affect the number of casing strings. Within a play, well
depths are often the most variable characteristic for casing with ranges of up to 5,000 feet. Operators
may also choose to run several casing strings to total depth or run a liner in lieu of the final casing string.
Casing costs can range from $0.6 MM to $1.2 MM, making up 9% to 15% of a wells total cost.

(3) Frack pumping costs can be highly variable, but are dependent on horsepower needed and number
of frack stages. The amount of horsepower is determined by combining formation pressure, rock
hardness or brittleness and the maximum injection rate. Pumping pressure (which includes a safety
factor) must be higher than the formation pressure to fracture the rock. Higher pressure increases the
cost. The number of stages, which often correlates with lateral length, is important since this fracturing
process, with its associated horsepower and costs, must be repeated for each stage. These total costs
(for all stages) can range from $1.0 MM to $2.0 MM, making up 14% to 41% of a wells total cost.

(4) Completion fluid costs are driven by water amounts, chemicals used and frack fluid type (such as gel,
cross-linked gel or slick water). The selection of fracking fluid type is mostly determined by play
production type, with oil plays primarily using gel and natural gas plays primarily using slick water.
Water sourcing costs are a function of regional conditions relating to surface access, aquifer resources
and climate conditions. Water disposal will normally be done by re-injection, evaporation from disposal
tanks, recycling or removal by truck or pipeline, each with an associated cost. Typically about 20-30
percent of the fluids flow back from the frack and require disposal. Operators typically include the first
30-60 days of flow back disposal in their capital costs. These costs can range from $0.3 MM to $1.2 MM
making up 5% to 19% of wells total cost.

(5) Proppant costs are determined by market rates for proppant, the relative mix of natural, coated and
artificial proppant and the total amount of proppant. Proppant transport from the sand mine or factory
to the well site and staging make up a large portion of the total proppant costs. Operators use more
proppant when selecting less costly proppant mixes, which are comprised of mostly natural sand as
opposed to artificial proppants. A higher mix of artificial proppants has often been used for very deep
wells experiencing high formation pressures. Overall the amount of proppant used per well is increasing
in every play. These costs can range from $0.8 MM to $1.8 MM making up 6% to 25% of wells total
cost.

11
EIA UPSTREAM COST STUDY

D. Evolution of Costs during the Past Decade


Markets and their Drivers Cost Indexing

Cost indexes show the relative costs of equipment and services over time (Figure 2-5). This analysis
assumes an index value of 1 for the cost of a given item during 2014. Future and historical increased
rates will be greater than 1, whereas lower rates will be less than 1.

The price spike for


casing in 2008 was a
result of increased
global demand for
steel while there was a
temporary steel
shortage. From 2010
to 2012 the industry
expanded faster than
the services and tools
industries could keep
up with, thus driving
up costs rates,
Figure 2-5: Historical nominal indices of key cost drivers primarily for frack fluid
volume, water disposal
and frack pumping units. As these services increased to meet demand, their costs decreased
significantly. From 2012 onward, improvements were also made to other services related to well
completions, such as additional water treatment plants, injection sites, proppant mines, more efficient
fracks and more experienced personnel. As a result, cost rates receded for some items and dropped
even faster moving into 2015. Further depressing the tools and services markets today are low oil and
natural gas commodity prices, which are causing drilling and completion activity to wane.

As Figure 2-5 shows, supply shortage is inelastic in the short term. Sharp increases in activity, where
essential services are in short supply, will spike costs until one or more of the following occurs: (1) the
cost increase has stifled the development pace enough to bring supply and demand back into balance,
thus forcing the service provider to lower its rates, (2) new methods are employed to avoid the cost; or
(3) an expansion of supply eventually catches up with demand as observed during the 2012-2014 period.
An example of new methods being employed relates to the first wells drilled in the shale plays, which
were completed primarily with completion rigs. Over time the completion practice evolved to the use of
coiled tubing, which was a response to increasing completion rig rates, as well as a response to slow
completion times, as coil tubing speeds up the completion process. During 2014 the market had
achieved a balance between supply and demand for most services. However with the drop in oil prices
and consequent drop in wells being drilled and completed, there is an over-supply of oil-field services.
This sharp contraction in demand is expected to lower prices significantly for many services as we will
discuss in this study.

Services in each of the plays experienced similar shifts in cost rates as many of the cost items, such as
proppants and oil field tools and tubulars, were able compete across multiple plays. Play specific-cost

12
EIA UPSTREAM COST STUDY

changes are related to services that are more regional in nature, such as rigs, water and pumping units,
which are not typically moved over long distances between plays.

Changes in Well and Completion Design and Application of Key Technologies

Over the past decade specific changes in technology have been employed to both reduce costs and
increase production. While costs may go up, the resulting performance benefit far outweighs the cost.

Technology improvements related to drilling:

Longer laterals (increase performance).


Better geosteering to stay in higher-producing intervals (increase performance),
Decreased drilling rates (decrease cost),
Minimal casing and liner (decrease cost),
Multi-pad drilling (decrease cost), and
High-efficiency surface operations (decrease cost).

Technology improvements related to well completion:

Increase amount of proppant superfracks (increase performance),


Number and position of frack stages (increase performance),
Shift to Hybrid (cross-link and slick water) fluid systems (increase performance),
Faster fracking operators (decrease cost),
Less premium proppant (decrease cost), and
Spacing and stacking optimization (increase performance).

Applying each of these factors leaves a footprint on increased capital efficiency, yet the specific effect of
each is difficult to measure, particularly against the backdrop of geological influences that also have a
profound influence on cost and
performance. Nevertheless, the
cumulative results, as discussed below, are
outstanding.

Lateral length: While this study focuses


primarily on horizontal drilling, we
acknowledge that the shift from vertical to
horizontal wells is the most important
change to occur over the last decade,
allowing for greater formation access while
only incrementally increasing the cost of
Figure 2-6: Historical drilling trends the well. Over the past decade, lateral
lengths have increased from 2,500 feet to
nearly 7,000 feet and, at the same time, we have seen nearly a three-fold increase in drilling rates
(feet/day) (see Figure 2-6). This increase in efficiency is leading to overall downward pressure on drilling
costs for each well, even though lateral lengths may be increasing.

13
EIA UPSTREAM COST STUDY

Completions: Within each play, larger amounts of proppant, fluid and frack stages are being employed to
drive up production performance (Figure 2-7). We also note that cheaper proppant and slightly less
water per pound (lb.) of proppant are being used to combat costs. With the well completion schemes
evolving and growing over time, we would
expect performance to also increase.
Average stage length has decreased from 400
to 250 feet which allows more proppant to be
used.

Often, only a few operators will use a


particular cost saving or production
performance improvement technique. As
others observe success with the new
techniques, they will often adapt it to their
well and completion design. For example,
the use of more, lower cost proppant was
initially attempted by only a handful of operators in the Bakken, but is catching on and is becoming the
preferred completion method in the play. Similarly, we would expect a continued evolution of well
design in the future as operators look for ways to become more efficient in an environment of lower oil
prices.

Multi-well pads and higher surface operation efficiency: Multi-well pad drilling allows for maximization
reservoir penetration with minimal surface disturbance, which is important in areas that are
environmentally sensitive, have little infrastructure, or in mountainous areas with extensive terrain
relief. Operational costs are reduced as this allows operators to check wellhead stats (pressure,
production, etc.) on numerous wells in the same location. Most pads are situated with 4 - 6 wells, but
some are planned for 12, 16, or even 24 wells where there are multiple stacked zones. With the surface
locations of wells on a pad being close to each other, mobilizing rigs from one well to another is also
more efficient. Walking rigs, automated catwalks, and rail systems allow rigs to move to the next
location in hours, not days. Facilities can be designed around pads, thus further reducing costs.

Improved Water Handling: As


water resources become more
and more scarce, operators are
being forced to come up with
better solutions for the amount
of water used for each well,
especially in arid regions, such as
the Permian Basin and the Eagle
Ford in South Texas. This is also
important in environmentally
sensitive areas. Many companies
are using recycled water for
drilling and completion
operations instead of having
Figure 2-8: Change in historical well cost comparison

14
EIA UPSTREAM COST STUDY

water trucked in or out. Using recycled water also reduces operators costs. For example, Apache was
paying upwards of $2.00 per barrel to dispose of water in the Permian Basin, but pays only $0.17 per
barrel to recycle.
Combining Indexing and Changes in Well Design to Track Historical Well Costs

Historical changes in overall well and completion costs can be attributed to changes in cost indices, as
well as changes in well design parameters. Figure 2-8 shows both the effect of well design and indexing
on total well costs:

Avg. Capex, Actual The Avg. Capex, Actual is the average total nominal well cost for each year
as it actually occurred. Note that overall costs are actually coming down, despite more complex
well designs of recent years. However a well still costs more in 2014 than 2010.
Capex for 2010 Cost Rates, Well parameters of the year This is comprised of the 2010 cost
rates being applied to the average well design of a given year. Note that had we held 2010 rates
steady, the actual cost of a well drilled in 2014 would have gone up slightly. If cost rates had not
come down since 2010, well costs would have grown by 40% due to the longer laterals and
increased use of proppant.
Capex for 2010 Well Parameters, Cost Rates of the Year This is comprised of well parameters
of 2010 with cost rates for the given year being applied. Note that the more simple well design
of 2010 would have cost about the same in 2014 when applying yearly index rates, but would
have cost much less than the more complex well design of 2014.

When a back-costing exercise is performed


we see a similar story unfold within each
play, as a well with a 2014 design drilled
back in 2010 would have cost roughly the
same (see Figure 2-9). Between 2010 and
2012, well cost rates increased along with
well dimensions and completion intensity.
This exacerbated the increases in well cost,
but improvements to efficiency and
improving well services and tools markets
since 2012 have helped overall well costs
come down since then.
Figure 2-9: Historical comparison of cost using
current well design

Overall Trends by Major Cost Component

Recent drilling costs make up a much smaller portion of total well costs compared to prior years for all
plays, as shown in Figure 2-10. This is due both to the growth in completion programs and associated
costs as well as efficiency gains, such as the drilling penetration rate improvements.

15
EIA UPSTREAM COST STUDY

Casing programs have been constant since


play inception, as geology and total depth
dictate their use and the most efficient
designs are determined as the first wells are
being drilled. Tubular costs, as a
percentage of total well cost, peaked during
2008 when there was a steel shortage in
the global market. Shortly after 2008,
casing rates dropped, while the increases in
other cost drivers have made casing costs
much less significant than in the past.

Frack pumping costs in 2015 have been


reduced in most plays down to 2010 levels
Figure 2-10: Contribution of drilling and casing despite much larger completions with more
stages. Nominal rates have dropped by
over 40% from their high in 2012, while the number of stages has increased from an average of 20 to 25.

As proppant amounts have grown, their contributions to costs have increased in importance when
determining total well cost in all plays except the Bakken (contribution to total well cost in the Bakken
has been variable from year to year). The Eagle Ford has also seen more expensive proppant mixes
used each year, making proppant costs much more important today than in prior years.

Fluid cost contributions were the greatest in 2012 when cost rates were highest. Since then, the rates
have come down by 60%, and fluid costs have contributed far less in recent years despite increased fluid
amounts currently used. The addition of gel use in some instances impacted total fluid cost, but even
this was overcome by improved cost rates.

Evaluating Effectiveness of Completion Design, Overall Trends in Cost/Boe

While additional well completion complexity


has increased costs, the aim of operators is to
reduce capital unit costs ($/Boe) needed to
develop the hydrocarbons by substantially
increasing the production performance. This
has proved to be quite successful in the
Midland, and Eagle Ford plays. In contrast, the
Bakken and Delaware have not substantially
improved; with unit costs remaining flat (see
Figure 2-11). In these instances, the goal of
increased completion complexity may be just
to maintain the current unit costs, as there are
a number of factors that can degrade
production performance, such as tighter down
Figure 2-11: Historic capex unit costs ($/boe) by play spacing or less desirable prospect selection.

16
EIA UPSTREAM COST STUDY

E. Future Cost Trends


Expected Cost Reductions

Oil prices, which had recently made a modest recovery, once again took a nose dive and, consequently,
IHS revised its oil price and production outlooks downward. WTI will remain below $45 for most of the
remainder of 2015 and will rise only slightly during 2016 (see Figure 2-12). Root causes underlying this
reduced forecast include:

High US and OPEC


production levels,
The return of Iranian oil to
the world oil markets, and
Weak demand growth
worldwide, particularly in China.

Consequently, oversupply will


continue for the next 12 months and
narrow in the second half of 2016.
Forecasted lower production (see
Figure 2-13) will result primarily from
an extended cut back in drilling, and
could become even deeper if prices
Figure 2-12: IHS historical and forecasted oil prices fail to recover.

This has led to a downward trajectory in costs. In 2015, total well costs will drop by 15% - 18%, on
average, from 2014 levels and are expected to drop another 3-5% in 2016. The dramatic drop in oil
prices has precipitated a huge
reduction in drilling and completion
services fees. During the third
quarter of 2014, which is the period
that this cost analysis represents,
there were approximately 770 rigs
actively drilling in the four plays. Over
the next several months this count
plummeted to only 350 as of July
2015 (Figure 2-14). Prices, which are
currently at under $50/bbl, are
expected to go lower, and IHS does
not anticipate a price recovery to
Figure 2-13: Revised production projection begin until mid-2016. World-wide
production levels are still out of
balance with demand expectations, and the higher cost U.S. unconventional plays will bear the brunt of
reductions in production as the markets seeks a new balance between supply and demand. This means

17
EIA UPSTREAM COST STUDY

that rig counts will fall even farther, resulting in continued downward pressure on costs for drilling and
completion services.

Primary cost drivers

Services such as pumping equipment and


specialized drilling rigs with 1,000 to
1,500 horse power (Hp) are primarily
used for unconventional play
development. Supply of these services
has expanded in recent years to
accommodate the high industry activity;
thus there is currently a huge supply
overhang, which will continue for several
years until prices recover to higher levels.
Some service companies are even
Figure 2-14: Monthly rig count by play expected to operate at a loss just to
maintain market share and keep their
skilled labor. As we anticipate cost reductions, we see the following rate changes for the five primary
cost drivers (see Figure 2-15):

Rig rates and rentals These services were created specifically for unconventional oil and
natural gas development. Thus, we expect to see reductions of 25 - 30% in 2015 from 2014
levels, with an additional 5 10% reduction in 2016, after which we would begin to see
increases of 5% during 2017 and 2018.
Casing and cement Casing cost is driven primarily by steel prices, which are expected to drop
by about 20% in 2015 due to general economic softness.
Frack equipment and crews
Like rigs and rig crews, these are
specialized for unconventional resources
and no other markets currently exist for
these services. We expect reductions
similar to those of rig rates and rentals.
Proppant We see reductions of
20% in proppant costs. The majority of
the proppant cost is due to transport
from sand mines in Wisconsin and
regional staging costs. There is little
room for further cost reduction here.
Frack fluids and water disposal -
Water sourcing costs are tied to
regulatory conditions and are not market
Figure 2-15: Projected cost indices of key cost based, although we expect large cost
drivers reductions in the cost of chemicals and

18
EIA UPSTREAM COST STUDY

gels. Disposal costs will not be affected by industry activity as rates are based on long term
contracts that escalate each year at around 1.8%. These factors may actually pose risks which
could drive costs up.
Other cost items will only see small cost reductions in the 5% to 10% range.

Future Well Design Trends

In a lower cost environment, continued emphasis will be made on gaining efficiencies and improving
performance in order to drive down unit costs ($/Boe). Attributes of well design will become more
interdependent and will continue to evolve as follows:

Drill days - Drilling gains are ongoing and are projected to increase into 2015. Normally, we
would have expected this to have leveled off by now, but drill bits continue to improve as
evidenced by the increase in drill feet per day. More pad drilling will decrease rig movement
times for mobilization and de-mobilization.
Lateral length Annual rates of increase are slowing, which may be due to limitations imposed
by lease and drilling unit size and configuration. Within a given drilling unit, operators will drill
their longest laterals first and then fill in the gaps with shorter laterals.
More proppant per foot Operators continue to push the limits as shown in Figure 2-16.
Production may continue to
increase as some operators are
using as much as 2,000lbs/ft. An
increased amount of closely
spaced wells are projected as
operators continue to harvest as
much of the resource as
possible. The extra proppant is
likely to be needed in order to
achieve the recovery rates
required for economic success in
these more closely spaced wells.
Nevertheless, some evidence
Figure 2-16: Historical trends of proppant (Lbs./Ft) exists that certain plays have
reached their maximum limit of
how much proppant can be used per lateral foot before well production is crowded out. This
may be true for the Marcellus and the Bakken where pay zones are typically thinner. As
proppant levels increase, additional fluid will be needed for emplacement.
More wells on a drill pad Facilities costs per well will decrease as facilities are increasingly
designed for the drill pad, not for the well. Other efficiencies such as water disposal, frack
staging and rig movements will also eat into costs.
Number of Stages - Operators are putting more frack stages within the lateral length as stage
lengths are decreasing to around 150-200 feet (with more closely spaced perforation clusters)
in order to accommodate the increased proppant amounts being used. Changing the
configuration is also improving production performance.

19
EIA UPSTREAM COST STUDY

Natural Proppants - Proppant amounts are expected to increase in all plays. However, proppant
types will move toward cheaper natural proppant, except in the Eagle Ford where proppant
mixes are becoming more weighted toward artificial sand.

Future Cost Projections

Each play will be affected differently by the changes in cost rates and well parameters going into 2015,
with savings ranging from 7% to 22%. Average well costs will be affected as follows:

Bakken well costs were $ 7.1 MM 2014, but will drop to $ 5.9 MM 2015.
Eagle Ford wells averaged $ 7.6 MM in 2014, but will be $ 6.5 MM in 2015.
Marcellus wells will be $ 6.1 MM in 2015 after having an average cost of $ 6.6 MM in 2014.
Midland Basin wells were $ 7.7 MM in 2014, but will drop to $ 7.2 MM in 2015.
Delaware Basin wells cost $ 6.6 MM in 2014 and will drop to $5.2 MM during 2015.

Additional cost decreases will occur in 2016, but by the latter half of that year we expect to see slight
recoveries in cost rates.

F. General Cost Correlations


The EIA is interested in projecting future costs by applying the parameters used, and therefore
correlations between major cost drivers and the actual costs within each play need to be understood.
Included within the discussion of each play is: (1) an analysis of the correlations of the well attributes
associated with the major cost drivers to the actual cost of that portion of the well, and (2) a comparison
of total well costs based on primary factors such as depth, amount of proppant and activity index (e.g.
cost per foot).

Correlation of well attributes

For this analysis we calculated costs by multiplying specific well design factors by specific rates to
determine the cost of each item. Total well cost was obtained by the sum of all of these subordinate
costs. As mentioned in Section C above, we then identified the top drivers that contribute to the overall
well cost and the contributing costs within each of these drivers; these are listed as follows:

Pumping Units for Fracking


o Injection rates (barrels per minute),
o Formation break pressures (psi), and
o Number of stages.
Drilling
o True vertical depth (TVD - feet),
o Lateral length (feet), and
o Drilling penetration rate (feet/day).
Proppant
o Amount of proppant (lbs.), and
o Cost per lb. of proppant (refers to the mix of natural and artificial proppants).
Frack fluids
o Amount of fluid (gallons),

20
EIA UPSTREAM COST STUDY

o Amount of gel (lbs. per gallon of water), and


o Chemicals (gallons per gallon of water).
Casing and cement
o TVD (feet),
o Lateral length (feet), and
o Number of casing strings.

The methodology for determining correlations between well design attributes and their associated costs
is described as follows. For each attribute: (1) we determined a range of well design inputs for 2010
through 2015 (using well data distributions and other applicable information) and projected these
ranges through 2018; and (2) calculated P10, P25, average, P75 and P90 values for each year from these
data distributions. We then applied the rates for each well design input to calculate a total cost for that
well design input. By comparing well design inputs with the resulting costs, an R-squared value was
generated based on the correlations between each P value and the resulting P cost for each
attribute. The results of this analysis will be presented for each individual play.

Total well cost per unit

We have demonstrated that there is a


strong correlation between well size,
complexity and costs. Also, we note
that the recent large declines in cost are
due to a drop in activity. This decrease
is partly due to an oversupply of rigs and
service providers, but may also be a
function of reduction in the number and
amount of services being performed.
For each play we will provide over time
Figure 2-17: Drilling cost rate per foot the following unit costs as based on
the following relationships.

Total Drilling Cost

Cost per foot


Cost per activity index

Total Completion Cost

Cost per unit of proppant


Cost per break pressure
Cost per stage
Cost per activity index
Figure 2-18: Completion cost rate per lb. of proppant Figures 2-17 and 2-18 portray play level
comparisons for simple unit costs. Drilling
unit costs per foot are the highest in the Midland Basin and lowest in the Bakken, while completion unit
costs per lb. of proppant are highest in the Bakken and lowest in the Marcellus. These figures also

21
EIA UPSTREAM COST STUDY

illustrate that while unit costs fall within relatively narrow bands for each play, other factors also
influence costs as well. Thus relying entirely on a single relationship to determine total cost is likely to
be misleading.

G. Key Take-Aways
In the current longer than expected low commodity price environment, operators face the challenge
of improving project economics and maintaining production growth at the same time. The demand
for new technology to bring the cost down is important; however, the majority of cost savings have
resulted from operators negotiating better rates with service providers.
Cost reductions have been occurring since 2012 as the supply of rigs and other service providers,
such as fracking crews, grew to meet the demand for these services. This cost reduction was
accelerated in 2015, when massive reductions in drilling resulted in a vast over-supply of services
relative to the demand.

Increased technology: Many advances in technology, such as geosteering, higher proppant


concentrations and closer spaced frack stages are increasing the overall cost of wells. However,
increased performance lowers the unit cost of production, which more than offsets the increased
expense of applying this technology.

Increasing efficiency: Service companies are seeing increased pressure from E&P companies to
reduce costs and improve efficiencies. For example, the number of drill days has decreased
dramatically in each play.

Operating Costs: The high variability of operating costs for lease operation, gathering, processing
and transport, water disposal and G&A offers operators an opportunity for cost reductions in the
future.

High-grading the production portfolio: Companies are adjusting capital spending toward the
highest-return elements of their asset portfolios, setting aside their inventory of lower-return
development projects until prices recover and/or costs decline sufficiently to move project
economics above internal hurdle rates. This trend is perhaps most pronounced in the US Onshore
shale plays.

22
EIA UPSTREAM COST STUDY

III. Deep Water Gulf of Mexico


Unlike onshore unconventional projects with massive manufacturing development, each offshore
project has its unique design and cost profile, including significant costs like dry hole costs.
Furthermore, there are fewer than 100 wells, including both exploration and development drilling, each
year in GOM deep water area, as opposed to several thousands of wells drilled annually onshore.
Although the number of
activities is much less than
onshore, the amount of
capital and time invested in
deep-water GOM is
comparable to onshore.

With fewer wells and much


higher costs, the statistical
well approach applied to
onshore unconventional
wells simply does not apply
to deep water fields.
Furthermore, the high
Figure 3-1: Phases of an offshore E & P cost cycle degree of specialization and
technical challenges of
offshore development and long development cycles has prevented the standardization of offshore
development and cookie cutter approaches.

A successful discovery and typical project will pass through a number of stages which will require
appraisal, development and production. Depending on various factors, such as water depth, size and
reservoir depth, a development concept is selected and development wells are drilled either before or
after platform or tie-back installation. Before production can begin, a hook up or construction of
infrastructure has to occur (see Figure 3-1). Each of these steps incurs significant capital expenditure.

A. Deep Water reserves, economics and oil price


Deep water drilling and production involves long-term, multi-billion dollar projects that take several
years to complete and are less impacted by short-term fluctuations in oil prices. Offshore operators
often have major project budgets for years and most projects are completed with the anticipation of
higher oil prices in the future. However, longer than expected low commodity prices have begun to take
a toll on GOM drilling. The industry faces the challenge of managing costs and encouraging
collaboration. Nevertheless, the rest of 2015 will continue to be driven by a combination of caution and
capital constraints. United States GOM activities will be heavily influenced by the perception of medium
term and long term oil prices, and any changes in activity levels are expected to lag significantly behind
that of onshore unconventional plays.

23
EIA UPSTREAM COST STUDY

Core plays in the Deepwater US GOM include the Plio/Pleistocene, Miocene, Miocene sub-salt, Lower
Tertiary, and Jurassic.
Future US GOM deep
water production
growth will come
primarily from three
playsthe Miocene
sub-salt, the Lower
Tertiary, and the
Jurassic. Each of these
three growth plays
offers different
opportunities based on
a companys risk
tolerance, skill set,
materiality
Figure 3-2: Creaming Curve (reserve additions) by deep water play
requirements, and
available capital. Nevertheless, since 2004 approximately 13,500 MMBoe of newly discovered reserves
in these plays is either being developed or is awaiting development (Figure 3-2).

As compared to other growth plays in the deep-water GOMthe Lower Tertiary and the Jurassic
development of the Miocene sub-salt has advanced because of the proximity to existing infrastructure,
which facilitates a lower commercial threshold to resource development, and more rapid development
of resource discoveries. On the other
hand, the largest growth plays from a
volume perspective (the Lower
Tertiary and the Jurassic) face
challenges in a sustained low oil price
environment due to constrained
commerciality caused by deeper
water depth and lack of
infrastructure (Figure 3-3). Most of
the Lower Tertiary and Jurassic fields
are over 150 nautical miles (nm) from
the shore and are well outside the
Figure 3-3: GOM deep water play boundary
extensive existing pipeline
infrastructure and platform. From a forward Figure
looking2-3: Primary cost
perspective, andrivers
assessment of IHS modeled US
GOM deep water sanctioned projects with estimated start dates between 2015 and 2021 reveals that a
majority of projects have an estimated forward development wellhead breakeven price below $50/bbl.
However, evaluating full cycle economics, the majority of the projects breakeven prices are above
$60/bbl, which puts unsanctioned projects at a great risk of cancellation or suspension. In a sustained
low oil price environment, companies must control costs, increase efficiencies, and access improved
technologies to further improve the economics in the larger frontier growth plays.

24
EIA UPSTREAM COST STUDY

Figure 3-4 is a comparison of estimated costs across


the three US GOM deep water growth plays (based
on 2014 cost environment) and shows that the Lower
Tertiary has the highest overall costs on a per barrel
basis. The Jurassic play has more favorable costs on a
per Boe basis than the Lower Tertiary due to a
slightly higher than average field size and better well
productivity. The Miocene sub-salt has smaller fields
and lower development costs, which stem from high
well productivity and proximity to existing
Figure 3-4: F&D cost by play infrastructure.

When studying the full cycle project economics, after taking into account operating cost and the fiscal
system under the late 2014 cost environment, most of the deep-water U.S. GOM current and future
projects are forecast to be uneconomic at oil prices below $50/bbl. However, from a forward
development perspective, most of the current US GOM deep water projects will go forward as a
significant amount of capital has been invested and operators are vigorously renegotiating their
respective contracts to secure the lower rates.

Furthermore, as part of the response to a lower commodity price environment, many of the large
operators in the deep water U.S. GOM have been revisiting development options and scenarios, with a
near-term focus on leveraging existing production infrastructure to develop discovered resources via
lower subsea tieback development costs. Infrastructure options tend to flourish within the conventional
Miocene deep water play; however, in more remote areassuch as the Lower Tertiary playthe
relative scarcity of production hubs and infrastructure provides fewer tieback options, which can act as a
constraint to field development.

B. Deep Water Cost Overview - Drilling


Each GOM deep water discovery has its own set
of features which influences the development
scheme and costs, ranging from geology, field
size, water depth, proximity to other fields,
reservoir depth and pressure, hydrocarbon
product, to operator preferences. The typical
development scope in the GOM deep water
includes the following: drilling and completion,
field development (which is primarily related to
the equipment and infrastructure installation,
such as production platform installation and
subsea tieback, platform construction and Figure 3-5: Water depth and well depth by major play
float over), and pipeline layout. Figure 2-3: Primary cost drivers
Well depth, reservoir quality, productivity, water depth and distance to infrastructure are key drivers to
drilling and completion costs. Of the three major plays, both water depth and well depth are shallower

25
EIA UPSTREAM COST STUDY

in the Miocene area (Figure 3-5), and thus the Miocene has the advantage over the other growth plays
due to its higher estimated well productivity
and relatively shallower reservoir depth
(20,000 - 24,000 feet SSTVD). The average
drilling and completion for Miocene wells is
approximately $120 MM (Figure 3-6); however
Miocene subsalt costs could be much higher,
given the complex geology and
unpredictability of the play.

The Lower Tertiary has experienced the most


technical challenges due to the combination of
water depth, well depth, high temperature, Figure 3-6: Well cost by major play
high pressure, and geological features of the Figure 2-3: Primary cost drivers
subsalt. Therefore, it inevitably experiences higher well costs. Jurassic projects are located in the
deepest water depth, which results in the highest well costs of the GOM at about $230MM (Figure 3-6)

C. Deep Water Cost Overview Development Concept


Each GOM deep water field discovery has its own set of features which influences costs, including field
size, water depth, proximity to other fields, reservoir depth and pressure, hydrocarbon product, and
operator preferences.

There are two types of field development in deep water: (1) standalone development and (2) subsea
development. The deep water wells are either developed through standalone infrastructure, a floating
production platform, or through subsea systems that tie-back to a production platform.

Since 2004, 35 deep water floating production platform systems (FPS) have been built and deployed in
the GOM deep water, bringing the total
to over 50 deep water production
infrastructures (Figure 3-7) in the GOM.
Tension Leg Platform, Spar, and
Semisubmersible are three major types
of floating facilities that perform
processing and handling of production
from deep water fields. Only one
Floating Production Storage Offloading
system (FPSO) is currently deployed in
the GOM by Petrobras because of
unfavorable regulation preference from
the Bureau of Ocean Energy
Management (BOEM).
Figure 3-7: Current and future hub facilities
Water depth, production capacity, hull
design, and topside design, including processing equipment and utility module, and drilling capability
drive the cost of these floating facilities. The majority of facility hulls have been built in shipyards

26
EIA UPSTREAM COST STUDY

overseas, mostly in South Korea, Singapore, and Finland, to minimize construction costs. Nearly all
topsides, on the other hand, are still built in the US as their technology is extremely complex.

Subsea production systems are applied in two scenarios: (1) they connect smaller fields to nearby
existing infrastructure; or (2) they can be applied to an area where existing infrastructure is scarce,
especially in emerging plays such as the Lower Tertiary and Jurassic.

Given the low oil price environment and the significant amount of deep water discoveries, the operators
have widely adopted the hub concept, which includes several jointly developed fields, with a center
floating production infrastructure to handle and process hydrocarbon product through flexible riser and
subsea tie-in. The Perdido project, which went online in 2010, was the first Lower Tertiary hub brought
on stream, followed by Cascade/Chinook in 2012 and Jack/St. Malo in 2014. These hubs, with the
addition of the Miocene Sub-salt Lucius hub (which is expected on stream in early 2015), could provide
proximity to infrastructure and accelerate the development in those frontier areas.

While breakeven prices vary across


projects, Figure 3-8 shows the estimated
average full cycle wellhead breakeven
price by play and development concept
at 2014 cost and price environment. It
demonstrates that the majority of Lower
Tertiary reserves have a breakeven
higher than $60/bbl. Meanwhile, the
greatest portions of the modeled
reserves for the Miocene sub-salt play
have a breakeven price below $60/bbl.
Monetization is a greater constraint for
those growth plays in more frontier
areas of the deep water basin. The
Jurassic and Lower Tertiary plays are
located farther away from existing
pipelines and platforms than the
Miocene sub-salt. This constraint is
expected to diminish over time as the Figure 3-8: Wellhead breakeven price by play and
plays mature and production hubs are development concept
established in currently frontier areas. In the Jurassic play, the semi-sub development of Appomattox
requires a ~$60/bbl oil price to break even. However, the tieback of Vicksburg to Appomattox requires
only a ~$48/bbl oil price to break even.

D. Cost Outlook
Our outlook is for a 15% reduction in deep water costs for drilling and related services in 2015, followed
by a marginal average increase of 3% per annum in overall deep-water costs from 2016 to 2020. This
cost deflation is material in many areas impacting deep water costsbut particularly in the rig market,

27
EIA UPSTREAM COST STUDY

where a rig overbuild long forecast for 201516 is now colliding with reduced demand, and resulting in
highly reduced rig day rates.

E. Key Take-Aways
At the current longer-than-expected low commodity price environment, the GOM deep water operators
face a tremendous challenge on cost savings and are striving to attain a balance between improving
project economics and maintaining production growth at the same time. The demand for new
technology to bring down costs and to improve productivity has hit an unprecedented high level,
especially in ultra-deep water and technically challenging areas. There are several initiatives that have
been proposed and discussed in the deep water industry:

Deferring unsanctioned projects: While capex cuts have reduced scope for spending on development
projects, the largest impact is likely to be felt on those projects which have yet to be sanctioned.
Conversely, projects already sanctioned and under construction are less likely to be delayed or
cancelled. However even in this case, the potential for deferral will increase should oil prices continue to
languish.

Reducing exploration capex: Several companies have focused capex cuts on their exploration budgets,
including ConocoPhillips, Marathon Oil, Murphy Oil, and TOTAL. If sustained, such a trend could have an
impact on longer-term production profiles via a reduced ability to restock development portfolios and
replace reserves. On the other hand, lower exploration spending will have little impact in terms of
reduced production growth over the near to medium term.

Increasing efficiency: Service companies are seeing increased pressure from E&P companies to reduce
costs and improve efficiencies. To the extent that operators and their partners can be successful in this
endeavor, E&P companies may still have the ability to proceed with key projects but at reduced levels of
investment.

Industry standardization: Besides subsea standardization, which is the most talked about piece of the
cost saving puzzle, standardization of delivery schedule, procurement and maintenance could help lower
costs. Today, most of the operators use various equipment designs, which often change for follow-on
orders for a given project. The sporadic and unpredictable nature of these orders can add significantly to
project costs.

Sticking to the timeline: Delay due to changes in requirements mid-project is currently one of the
biggest drivers of lower returns on some offshore projects. This directly contributes to both cost
overrun and production startup deferral.

Subsea boosting technology: Lower oil prices have prompted the operators to evaluate the alternatives
such as installing subsea boosting systems on the sea floor for existing producing fields to improve
production recovery instead of pursuing new field development projects. The operators are forced to
face the dilemma of evaluating the economics of drilling new wells versus applying subsea boosting
pumps on existing wells, and as a result, subsea boosting technology has come back in the spotlight.

28
EIA UPSTREAM COST STUDY

IV. Methodology and Technical Approach


A. Onshore Basins
IHS took the following steps to prepare the cost estimates for the onshore basins:

Sub-play definitions

IHS defined sub plays for each basin or play by locating the geographic areas in each play that shared
similar depth ranges, hydrocarbon type (predominately oil or natural gas), depth range, and production
performance. For the Permian Basin, we selected the most active and productive unconventional oil
plays. Well costs and cost ranges were determined for each sub-play.

Calculating well costs for each sub-play

IHS determines onshore unconventional well costs using its North American well cost model, which was
developed over several years during the height of the unconventional shale revolution and represents
costs as of third quarter 2014. Costs are determined by creating a typical well design for each sub-play
and multiplying each cost item or parameter by a nominal unit rate:

Rates: IHS maintains a database which captures service and tool cost rates from each play in
North America.
Well parameters for each sub-play are determined from IHS well data for recent wells of 2013
and 2014 vintage belonging to the sub-play. For example some of these parameters include
vertical depths, horizontal lengths, casing programs, proppant amounts and types, fluid
amounts and types, and drilling days (See Figure 4-1 for detailed listing).
The costs for each item are then determined by multiplying the amount or number of units
pertaining to a well parameter by the rate.

Figure 4-1: Cost components used in the cost model to derive total well cost

29
EIA UPSTREAM COST STUDY

Operating costs consist of gathering, processing, transportation, and water disposal and fixed well or
lease operating costs, but unlike capex, these items are mostly determined by the locality of the play or
sub play and are a function of infrastructure, the need for processing and other contractual
arrangement between operators and providers. Each operating cost rate in the model is researched
based on reports by media and direct contact with operators and is captured at the play level.

Benchmarking Costs with Published Data

In order to ensure accuracy of cost estimates, IHS researches the total well costs and any other data
available from operator reports and investor presentations and compares it to the costs calculated by
the cost model. These reported comparisons are included in the detailed cost discussion for each play.

Key Cost Contributors or Drivers

After costs for each sub-play were determined, major cost-contributing drivers were determined by
grouping together some of the smaller capital cost categories in order to consolidate the analysis to a
more manageable level of 11 categories (see Figure 4-2). The five largest categories comprising
approximately 75- 78% of the total well costs and 81% of total drilling and completion costs (excluding
facilities) were selected for further analysis. The remaining cost attributes were grouped together into
other (see Figure 4-3).

Figure 4-3: Key cost drivers


Figure 4-2: Detailed well cost components

Range of Costs

Within each basin, play, and sub-play, well drilling and completion attribute data pertaining to each of
the five major cost categories or drivers was extracted for each well from the IHS well database. Data
analysis was performed on the distributions to calculate high cost (P10), low cost (P90) and arithmetic
averages of each attribute. Using rates from the cost model, a cost was assigned to each P10, P90 and
average attribute value. Since raw data was extracted from the database, filters were applied to remove
obviously anomalous data points and recompletion and sidetrack data that would have misled the
study. The P10 data points extracted represent the high cost well inputs and the P90 data points
represent the low cost well inputs.

30
EIA UPSTREAM COST STUDY

Additionally, the selection of P10 and


P90 data points was intended to cut
off what are expected to be outliers
in the data. Figure 4-4 is a typical
illustration of the well attributes that
Casing pertain to each of the five main cost
categories or drivers and their ranges
Frack within the total cost of each
Fluid category.
Proppant
Rig rental The extent to which a well
parameter drives costs is determined
Frack by how much the cost of a well, with
equipment the average characteristics, changes
when moving a single input to the
P10 and P90 values. This creates a
range of costs representing the
Figure 4-4: Attributes pertaining to each key cost driver
distribution for a given parameter.

Historical Costs

Determining historical costs is similar to the determination of 2014 costs, in that both nominal rates and
specific well parameters are determined and multiplied together to obtain the cost for each well
parameter.

To determine historical rates, IHS maintains nominal capital cost rate indices for onshore field
development in the CERA Capital Cost Index report describing historical changes to cost rates for general
items such as casing, cement, mud, rigs and labor. In addition, we have developed rate indices specific
to onshore North America unconventional wells for frack fluid chemicals, gel, frack equipment, proppant
and water. These historical rate indices are based on historical data provided through research, industry
contacts and manufacturers as well as reported drilling rig day rates and proppant costs per lb.

IHS also maintains an operating cost index similar to the capital cost index and we have supplemented
this with rates specific to North American unconventional wells. For each year beginning in 2006, these
historical indices are applied to historical well parameters to determine the cost for each attribute and
cost category for each year.

In order to determine historical ranges of cost, well attributes were captured from the data going back
as far as 2006 or at the beginning of play inception. The distribution of data for each attribute within
each given year was analyzed to determine the P10, P90 and average needed to determine historical
cost range for each year, in a similar fashion as described for the 2014 cost model and analysis.
Additionally, the IHS cost indices were applied to the 2014 cost model rates to create historical cost
distributions for each year. Combining the historical well parameters with the historical cost rates
historical well costs and their distributions were determined annually.

31
EIA UPSTREAM COST STUDY

Future Costs Rates

This study includes projections through 2018 of nominal capital cost rate changes with special focus on
the differences between 2014 costs, which were analyzed in detail, and 2015 costs, which are todays
reality given the recent collapse in oil prices. Onshore unconventional well cost rate forecasts rely on
insight developed through interaction and leveraging of analysis from its specialist legacy companies
such as PFC and PacWest, as well as identifying and projecting certain trends. Assumptions, described in
the onshore summary portion of this report were vetted with research peers to provide a view
consistent with other IHS outlooks. We assume that price forecasts for oil will remain low through mid-
2016 with only modest recoveries through 2018; this implies industry activity will continue to drop off
and may not fully recover in the near term, thus sharply reducing rig rates and frack crew rates, which
are two of the five cost category drivers being analyzed.

Future well cost trends were developed by noting salient changes over time in key well attributes, such
as proppant usage. These were combined with future cost rates to project costs into the future.
Forecasted well parameters assumed mostly linear trends given the last few years unless there was a
reason to assume otherwise.

Cost Effectiveness

In order to assess cost effectiveness, a relationship between well total cost and well performance was
required. To evaluate well performance, well production curves were developed for each sub-play in
order to calculate estimated ultimate recovery (EUR) for each vintage year beginning in 2010 through
2014 and thereafter forecasting EURs for wells to be drilled from 2015 to 2018 based on current trends.
Total well costs were divided by the respective EURs for each given year and sub-play to determine a
unit cost as $/Boe.

B. Offshore Deep Water


Deep water field development costs, with granular data describing each component of the project, are
difficult to obtain. This is unlike shale plays where applications for expenditure (AFE) and drilling and
completion (D&C) costs are often touted by operators for each of their respective plays. Offshore Deep
Water Gulf of Mexico data has far fewer wells and fewer operators to produce data, which is mostly
quoted at the project level without any breakout between D&C, infrastructure, installation/hookup, etc.
In order to shed light on the costs of deep water developments, IHS produces a field development
costing software Que$tor to provide the breakouts and estimate costs by component. Supplementing
this is industry media research and experience which provides confirmation of total cost and component
costs for some project models. However, due to market changes or cost overruns the reported and
estimated figures are subject to change. Que$tor provides a relatively reliable industry standard for cost
analysis and lends itself well to IHS capital and operating cost index forecasts when certain cost data is in
short supply.

32
EIA UPSTREAM COST STUDY

IHS Questor is a tool developed over the last 25 years by engineers for engineers in order to assist in
assessing and managing the potential cost of a field development for green fields. Used by more than
200 companies worldwide, it is designed for pre-FEED work and is able to produce full cycle costs within
35-40% without many assumptions other than development concept, reserves and a few commercial
parameters for distances to shore, etc. Que$tor is comprised of databases of field and reservoir
properties to provide expected values of parameters when data is unavailable and also contains a
detailed cost database for each component in a field development for everything from rigs to pipelines.
Field level and reservoir characteristics are sourced from the IHS EDIN E&P activity database which
documents all events and qualities of fields throughout their lives. The cost data that Que$tor uses
comes from industry reports and direct contact with operators, which means Que$tor costs are more or
less reflective of actual cost data. Que$tor then applies a series of algorithms using the field
characteristics and the relevant cost data to produce cost for each development component at a
granular level.

The unit cost database in Que$tor is based on Q3 2014 cost collection. For example, deep water rig day
rates, such as semisubmersible and drill ship representing GOM Q3 2014 contracts. The other unit costs,
such as Christmas tree, casing, tubing, cementing, logging, and wellhead equipment also reflect Q3 2014
cost. We selected five projects in GOM deep water representing typical reserve size, and field
development plan from three plays, Miocene super salt and subsalt, Lower Tertiary, and Jurassic. The
reserves, well depth, water depth, well productivity, reservoir pressure and temperature are plugged
into the Que$tor model to generate drilling and completion cost. Production platform costs are modeled
based on water depth, capacity, and the platform type. Subsea tieback and pipeline layout cost and are
also modeled based on the distance to the host platform and detailed field design.

33
EIA UPSTREAM COST STUDY

For forward looking cost estimates, we rely on the IHS rig rate forecast, capital cost index to forecast
future development cost.

For high-level play level breakeven prices, development costs, and regional development scenario
outlook, we use guidance from IHS Global Deepwater and Growth Plays Service, which is an analytical
research service providing play-level analysis of the commercial development of currently developing
resources, and highlighting both materiality and value potential in each play area.

34
EIA UPSTREAM COST STUDY

V. Bakken Play Level Results


A. Introduction and sub-play description
The Bakken oil play is located in the Williston Basin of North Dakota and Eastern Montana. Producing
formations include both the Bakken and Three Forks, which are fairly uniform throughout the basin and
occur at approximately 10,000 foot depths. Horizontal drilling began at Elm Coulee Field in the early
2000s and then moved to Sanish-Parshall in 2007 as that sweet spot was delineated. Two additional
areas, namely the New Fairway, emerged with unique factors that have influenced drilling and
completion costs (see Figure 5-1).

Figure 5-1: Location of Bakken Three Forks sub-plays


Drilling in the play has increased steadily since play inception (figure 5-2). Production ramped up quickly
to 1.2 MM barrels/day, but has leveled off as oil prices have plummeted. Rig counts that once exceeded
200 have fallen to the mid-eighties in recent months due to the oil price decrease. The play is located a
long distance from oil markets and have
limited infrastructure access to both oil and
gas markets due to the recent significant
production increases. Natural gas issues have
been mostly overcome by partial flaring of
excess associated gas, development of new
gas plants and gas take-away capacity. Oil
transportation has relied on rail for nearly
50% of oil production in order to reach
markets on the east and west coast.
Figure 5-2: Drilling history of each Bakken sub play

35
EIA UPSTREAM COST STUDY

B. Basic Well Design and Cost (2014)


Total Bakken Cost

Total well costs range from $7.5 MM to $8.1 MM as shown in Figure 5.3. The consistency in TVD, lateral
length, pressure and completion design amongst the sub-plays is reflected in similar costs amongst the
sub-plays for drilling and completion. Exceptions include the Elm Coulee field with lateral lengths of just
8,600 feet which are shorter and use less proppant, thus reducing completion costs, and the New
Fairway which has a greater TVD, and thus has higher drilling costs.

Comparison with Published Data

The average Bakken well cost of $7.8 MM compares with published costs reported by operators in 2014
as follows:

Operators reported costs range from MM$ 6.5 to MM$ 9.6 with Oasis reporting the lowest and
Continental reporting the highest.
EOG and SM Energy averaged over MM$ 9 with EOGs minimum being MM$ 8.
Hess and Halcon wells were approximately MM$ 8 with Hess achieving their lowest cost wells at
MM$ 7.2.
Elm Coulee - Continental reported costs of MM$ 7 to 8.5.
Periphery - Operators reported cost of MM$ 7 to 9.
Parshall - Operators reported cost of MM$ 6 to 8.
New Fairway - Operators reported costs between MM$ 7 and 9.6.

Figure 5-3: Total Bakken cost by sub-play

36
EIA UPSTREAM COST STUDY

General Well Design Parameters

Table 5-1 summarizes well design parameters for each sub-play. Proppant mixes, amounts and
horsepower drive costs, and we note that Parshall uses the most horsepower and proppant, but less
artificial proppant. Casing programs are uniform with a conductor pipe, two strings and a liner generally
used. Artificial lift soon after the well comes on stream is the common practice.

New
Well Parameters Unit Elm Coulee Parshall Periphery
Fairway
TVD Ft 10,069 10,169 10,905 10,030
Horizontal Ft 8,630 9,018.90 9,513 9,670
Formation pressure Psi 6,042 6,102 6,543 6,018
Frack stages # 25 30 30 31
Frack break pressure Psi 9,969 9,763 10,469 9,629
Pumping rate Bpm 50 55 46 45
Horse Power Hp 14,049 15,135 13,573 12,213
Casing, liner, tubing Ft 31,504 32,494 35,108 32,849
Drilling days Days 27 24 26 25
Natural proppant MM Lbs. 1.86 4.13 3.77 1.78
Artificial proppant MM Lbs. 1.86 0.46 0.42 1.78
Total Water MM gal 2.89 4.37 3.63 3.3
Total Chemicals Gal 144,497 218,649 181,413 164,968
Total Gel Lbs. 115,598 43,730 36,283 32,994
Table 5 1: Properties of typical wells in each sub-play used to calculate costs

Wells in Elm Coulee are drilled to just over ten thousand feet vertical depth and have lateral lengths
averaging 8,600 feet. The long lateral lengths are more than sufficient for large completions with 25
stages using over 3.7 MM lbs. of proppant and nearly 3 MM gallons of fluid. Proppant mixes here are
fairly expensive with a heavy use of resin coated mixed with natural sand. Completion fluids are nearly
always gel based, which is typical of oil plays.

Wells in Parshall are drilled to nearly 10,200 feet vertical depth and have lateral lengths over 9,000 feet.
Long lateral lengths support 30 stages using over 4.6 MM lbs. of proppant and nearly 4.4 MM gallons of
fluid. Despite using resin coated and ceramic proppant, mixes here are fairly inexpensive due to the fact
that they are heavily weighted to natural sand. Completion fluids are mostly gel with some wells
completed using slick water.

Wells in Periphery are drilled to over 10,000 feet vertical depth and have lateral lengths of nearly 9,700
feet. Long lateral lengths support 31 stages using over 3.5 MM lbs. of proppant and nearly 4.4 MM
gallons of fluid. Despite using few proppants for the large number of stages, proppant cost is high with
heavy use of ceramic sand. Completion fluids are mostly gel with some wells completed using slick
water.

Wells in New Fairway are drilled to over 11,000 feet vertical depth and have lateral lengths over 9,500
feet. Long lateral lengths support 30 stages using over 4.2 MM lbs. of proppant with over 3.6 MM

37
EIA UPSTREAM COST STUDY

gallons of fluid. Proppant cost is not high as wells use mostly natural sand and 100 mesh. Some wells
use ceramic proppant which would drive up the well cost significantly. Completion fluids are mostly gel
with some wells completed using slick water.

Within the Bakken operators use the sliding sleeve technique, instead of the traditional plug-and-perf
fracking procedure, for fracking wells while reducing completion costs.

C. Operating Costs
Operating costs are highly variable ranging from $15 to $37.50 per boe (Figure 5-4) and are influenced
by location, well performance, and operator efficiency.

Figure 5-4: Range of operating expenses

Lease Operating Expense (LOE)

Most of the Bakken lease operating expenses (LOE) incurred relate to artificial lift and maintaining
artificial lift. However, a few companies are able to nearly avoid most of these costs. Another major cost
for LOE is water disposal, as the Bakken produces 0.75 to 1.0 bbl. of water for every bbl. of oil that is
produced. Direct labor and other costs are fairly small relative to the rest of the costs, but are similar to
other plays. The Other category contains common costs like pumping, compression, and other recurring
types of costs, which are mostly determined by the cost of energy to run them (figure 5-5).

38
EIA UPSTREAM COST STUDY

Figure 5-5: Breakout of LOE costs

Gathering, Processing and Transport (GPT)

Oil is sent by either pipeline or rail to several destinations after being transported to a loading area. The
range of costs or differential incurred depends on whether transport is by rail or pipeline. Train
transportation is the only option for transport to the east or west coast and can cost $10-$13 per barrel,
while pipeline transport to the gulf can save much as $5 per barrel or more. The breakout of GPT costs
is presented in Table 5-2.

Gas has had very few market options, due to the fact the Bakken area was not as productive as other
regions of the U.S. during the major conventional field developments and pipelines are limited. Gas
plants and pipelines are being built, thus reducing gas flaring. As of 2014, gas was still flared for up to
30% of the wells. This activity is expected to result in 100% marketed gas in the near-term. Current gas
processing, fractionation and transportation rates are in line with other plays despite being limited in
availability. Access to markets is in fairly close proximity with destinations for product in Chicago,
Edmonton, and other northern markets.

Bakken Bakken
Units
High Low
Gas Gathering $/mcf 0.35 n/a
Gas Processing $/mcf 0.75 n/a
Short Transportation Oil $/bbl. 0.35 0.2
Long Transportation Gas $/mcf 0.25 n/a
Long Transportation Oil $/bbl. 12.50 6.25
Long Transportation NGL $/bbl. 12.50 n/a
NGL Fractionation $/bbl. 3.50 n/a
Water Disposal $/bbl. water 8.00 4.00

Table 5-2: Breakout of GPT costs


Figure 2-3: Primary cost drivers

39
EIA UPSTREAM COST STUDY

G&A Costs

G&A costs range between $2.00/ boe and $4 .00/boe. These may increase during 2015 due to layoffs
and severance pay-outs, but will be reduced over time due to staff reductions

Cost changes in 2015

Table 5-3 below summarizes operating cost changes that we expect to see between 2014 and 2015
going forward.

Item Change Description of change for 2015


Gas Gathering -4% Current contracts are sticky, but additional gas infrastructure will
allow for more gas to be marketed. This will increase the cost rate
for those who flare, but this will net a higher value.
Gas Processing -4% Current contracts are sticky, but additional gas infrastructure will
allow for more gas to be marketed. This will increase the cost rate
for those who flare will now pay this, but this will net a higher value.
Short Improved pipeline infrastructure will allow for less trucking.
Transportation Oil -3%
Short -5% Improved infrastructure will allow for more piping of production, but
Transportation Gas many operators will incur the same costs as 2014.
Long Transportation -10% Lower rail activity and improved infrastructure will drive this
Oil improvement.
Long Transportation -5% Improved infrastructure will allow for more piping of production,
NGL specifically a 5% decrease, but many will incur the same costs as
2014.
NGL Fractionation 0% No change expected.
Water Disposal +1.80% Many water disposal contracts have fixed rates. Some of this will
escalate based on PPI or another index. Only companies that dispose
of their own water will see savings
G&A +5% Severance package/payments due to layoffs are increasing G&A
despite lower future operating cost. Savings will not be realized until
2016.
Artificial Lift -10% Oil field services rates are dropping due to lower activity and lower
input costs rates like energy.
Artificial Lift -10% Oil field services rates are dropping due to lower activity and lower
Maintenance input costs rates. Maintenance will now be avoided in some cases
where it was profitable at higher prices. Companies that pay a fixed
maintenance may not see better rates in 2015 unless they are able
to renegotiate.
Direct Labor -3% Saving here will be due to fewer operational employees.
Other (pumping, -10% Energy costs savings.
compression, etc.)
Table 5-3 Changes in operating expense going forward

40
EIA UPSTREAM COST STUDY

D. Leasing Costs
Lease acquisition costs will depend on if the operator has secured acreage before the play has been de-
risked, as explained in Chapter 1. Figure 5-6 provides recent transaction costs per acre and the
incremental cost to each well that is incurred.

We are assuming that each lateral is going to


require 640 acres and that two stacked
laterals can be drilled, one in the Bakken and
the other in the Three Forks, for a net
requirement of 320 acres per well.
Approximately 10-20% of the acres acquired
will not be utilized. Ultimately we begin to
see that paying $6500/acre will add up to an
additional $2.5 MM per well.

Figure 5-6: Historical leasing costs

E. Key Cost Drivers and Ranges


Overall, 74% of a typical Bakken wells total cost is comprised of five key cost drivers (see Figure 5-7):

Figure 5-7: Bakken capex breakdown

Drilling:
o rig related costs (rig rates and drilling fluids) 17% or $1.32 MM
o casing and cement 11% or $0.86 MM
Completion:
o hydraulic fracture pump units and equipment (horsepower) 25% or $1.95 MM
o completion fluids and flow back disposal 11% or $0.86 MM
o proppants 10% or $0.78 MM

41
EIA UPSTREAM COST STUDY

Range of Costs and Key Drivers

Various cost attributes are classified within each of the five major key drivers as shown in Figure 5-8.
The total cost for each of these five cost drivers is portrayed with P10/P90 ranges created for each of
the contributing attributes pertaining to such range. These ranges are intended to portray variation
and uncertainty.

Pumping costs, which is the most costly driver, exhibits the most variation; suggesting that significant
deviation from the norm could add or decrease significantly from the total drilling cost. Injection rates
have a range of 31 bpm to 72 bpm, which has the largest effect on pumping costs creating a range of
MM$ 1.6; with possible cost increases over the average by $ 0.9 MM or cost decreases over the average
by $ 0.7 MM.

Drilling penetration rate


variability, from 411 Ft/d to
965 Ft/d, creates a drilling
cost range of $ 0.9 MM; with
possible cost increases of up
to MM$ 0.7 for wells that
drill slowly or cost decreases
of up to $ 0.2 MM for drilling
faster than the average.
Drilling penetration rates are
skewed toward faster
drilling, as it is actually quite
rare for a well to be drilled at
Figure 5-8: Range of cost for attributes underlying key drivers the slower end of the
distribution.

The proppant amount variability, from MM lbs. 3.5 to MM lbs. 12, creates a proppant cost distribution
of MM$ 1.7, with the potential to lower costs by just MM$ 0.1 and raise costs by MM$ 1.6. Most wells
use proppants at the lower end of range. The fluid cost range for total fluid amount is MM$ 0.7, with
the potential to raise costs over the average by $0.3 MM or lower them by $ 0.4 MM (with fluid
amounts ranging from 1.9 MM gallons to 5.3 MM gallons). The range of vertical depths in the play, from
9,263 feet to 11,147 feet, creates a casing cost range variation of just $ 0.1 MM. Upward or downward
cost movement in this category is negligible.

F. Evolution of Historical Costs


Historical Well Costs
Between 2008 and 2009 steel costs rose significantly. This created a spike in 2008 that was followed
shortly thereafter by a drop due to oil price decreases in the latter part of that year (see Figure 5-9).

42
EIA UPSTREAM COST STUDY

Between 2010 and 2012 nominal


well costs in the Bakken remained
under $ 5 MM until horizontal
development throughout the US
took off in 2011 and costs such as
rig rates and frack crew rates
began to rise. The 2011 and
2012 years saw huge price
increases, approximately $ 1.5
MM per year. Because of rising
rig rates, drilling costs have
increased despite improved
drilling efficiencies. Proppant and
Figure 5-9: Historical nominal well cost by major cost driver fluid costs increased 60% to 70%
and continued to increase year-
on-year. The number of stages, lateral length and increased proppant (with commensurate fluids and
chemicals) further fueled cost increases. With increased activity, water sources and disposal facilities
were limited. Along with greater numbers of stages, proppant and fluid, associated pumping costs grew
and were further exacerbated by shortages in completion service labor and equipment. Casing costs
have remained fairly flat throughout the entire period.

As the service industry grew to meet demand between 2013 and 2014, rates for pumping equipment
and fluids subsided and overall costs decreased. Nominal costs in 2013 dropped by about $1.0 MM, but
stayed fairly constant in 2014.

Changes in Well and Completion Design

Between 2006 and 2011, lateral length steadily increased until it reached its current length of just less
than 10,000 feet. On the other hand proppant per well has grown steadily year over year and feet per
stage has decreased more slowly. This suggests that fluid and proppant concentrations in each stage are
increasing (Figures 5-10 and 5-11). Despite downward pressure on rates from 2013 to 2014, the
additional proppant per well in year 2014 contributed to a slight increase in cost for a well.

Figure 5-10: Lateral length and total depth Figure 5-11: Proppant per well history
history

43
EIA UPSTREAM COST STUDY

The mix of frack fluids has


evolved over the years,
beginning with predominately
water fracks. Almost
immediately in 2011, operators
switched to X-link gels (Figure
5-12). At the same time
information gathering
improved. Well EURs increased
from 381 kBoe in 2010 to 544
kBoe in 2011. This suggests that
x-link gel fracks and additional
proppant were having a positive
impact on performance and
Figure 5-12: Change in frack fluid use over time that the additional capex was
paying off. Overall, play Capex
cost per Boe dropped from $13.24 per Boe in 2010 to $12.48 in 2011, which is the year that X-link gels
were first used. Since that time there has been some erosion in performance.

There have been recent decreases in lateral lengths, as it appears that 9,000-10,000 feet is the best
balance between cost and EUR. The overall decrease in average EUR from 451 kboe in 2011 to 391k in
2013 is likely due to drilling wells outside sweet spots due to higher oil prices. At the same time
efficiencies in drilling and completions have reduced costs from 2012 to 2013 (Table 5-4). In 2014, EURs
again began to increase and we see the trend continuing as operators become more selective in their
drilling locations due to lower oil prices.

Year $/Boe EUR -Boe


2010 15.79 298,129
2011 14.32 451,013
2012 19.06 407,423
2013 17.05 390,842
2014 15.67 425,627

Table 5-4: Drilling and Completion Unit Cost

Some of the performance increase may be due to operators applying larger and larger amounts of
proppant (see Figures 5-13 and 5-14). However this may not be as effective as hoped for, as the EUR per
unit of proppant is decreasing. In other words the amount of proppant used is increasing faster than
performance improvement. The evidence suggests that despite the use of improved technology the
performance increases have more to do with site selection. Furthermore applying technology will only
allow operators to tread water as they struggle to maintain performance and at the same time
attempt to reduce their costs per boe.

44
EIA UPSTREAM COST STUDY

Figure 5-13: Change in frack fluid use over time

Figure 5-14: Change in frack fluid use over time

45
EIA UPSTREAM COST STUDY

G. Future Cost Trends


Cost Indices

Bakken development activity is dropping sharply with little chance of recovery soon. Active rigs in the
play are in the mid-eighties and expected to drop into the mid-60s by the end of the year. Because the
Bakken is relatively infrastructure-constrained and much of the transport of oil is by rail, there is a huge
differential of $6-13 compared to
WTI. This will only further depress
activity. Also, being a regional
market for services, equipment
such as rigs and pumping units
will not be able to move easily to
other areas, thus putting more
pressure on service providers.
Overall, cost rates are decreasing
from 2014 levels by 20% during
2015, and will drop another 3-4%
in 2016 (see Figure 5-15).

Pumping and drilling rig cost rates


are dropping and are expected to
be 25 30% lower by the end of
Figure 5-15: Indices for major cost drivers 2015, with another 5% decrease
in 2016. Rates will begin to recover in late 2016, but will stay low through 2018. Proppant costs will
drop by 20-25% in 2015, largely due to decreases of 35-40% at the mine gates. The impact on fluid will
be less. Due to a forecasted drop of 20% during 2015 in the price of steel, tubulars, and other fabricated
materials will also cost less.

Changes in Well Design

Despite the challenging


environment operators will
continue to lower unit costs
($/Boe). The following trends are
expected to continue:

Lateral length - Average


lateral length has not
moved much during the
past four years and is
projected to remain
relatively constant at 9,200
feet (see Figure 5-16).
Vertical depths should also Figure 5-16: Historical and forecasted total depth
remain fairly constant.

46
EIA UPSTREAM COST STUDY

Stages - The average


number of stages is
projected to increase
from 28 to 32 in 2015
and by 2018 should
reach nearly 40 (see
Figure 5-17) and
because lateral lengths
are not projected to
change, we can expect
that stage spacing will
tighten considerably.
Drilling efficiencies -
These have already Figure 5-17: Historical and forecasted stages
been optimized and
any changes here will be small with average drillers achieving 780 Ft/d by 2018. This is up 10%
from 710 Ft/d in 2014 (figure 5-16).
Proppant - Proppant amounts will increase from 450 Lbs./Ft in 2014 to 550 Lbs./Ft by this year
and will steadily increase to 820 Lbs./Ft by 2018. This is still relatively light compared to the
1200-1400 Lbs./Ft we see in other plays (Figure 5-18). Proppant mix is expected to be focused
more heavily on natural proppants in order to afford more total proppant. Average fluid use is
expected to increase proportionately. Gel and chemicals used are expected to remain the
preferred option going forward as completion fluids types have been fixed for some time.

Figure 5-18: Historical and forecasted proppant

More wells being drilled on single drill pads As more wells occupy single drill pads we can
expect potential cost savings from shared facilities and other related items, such as roads, mud

47
EIA UPSTREAM COST STUDY

tanks, and water disposal systems. Of the total well cost, $1.1 MM is based on sharing costs
amongst four other wells. Table 5-5 illustrates how future drill pad configurations could save
money. For example, there are currently two stacked zones, namely the Bakken and Three
Forks, which are considered potential targets. Pilot programs have been completed for two
additional Three Forks zones, bringing the potential zones to four. Additional testing has also
been completed for tighter spaced wells, thus the potential exists for up to 16 wells to be drilled
from a single pad, which could save potentially $825,000 per well. These savings are not likely
to apply throughout the play, but will be focused more in localized areas. Nevertheless this
illustrates potential savings.

Stacked Distance Wells Cost of items


Horizons between per pad related to pad -
wells 2014
Modeled 2 3240 feet 4 $ 1,100,000 Modeled Cost
Traditional View 2 3240 feet 4 $ 1,100,000 Development Cost
Potential upside 4 1320 feet 16 $ 275,000 Potential New Cost
Difference 2 2 4 $ 825,000 Potential Savings
Table 5-5: Potential savings from additional wells being drilled from a single pad

Future Well Costs

Future changes in overall well and completion costs are quantified in forecasted indices, and are
combined with projections in future well design parameters to project future costs. Figure 5-19 shows
both the effect of well design and indexing on recent historical costs beginning in 2012 and future well
costs through 2018:

Figure 5-19: Comparison of actual future costs with forecasted

Avg. Capex, Actual This captures the average nominal total well cost for each year as it actually
is expected to occur. Note the acceleration of the rate declined in 2012, despite more complex
well designs of recent years which are expected to continue.

48
EIA UPSTREAM COST STUDY

Capex for 2010 Cost Rates, Well parameters of the year This captures the application of 2012
cost rates to the average well design of a given future year. Note that had we held 2012 rates
steady through the forecast period, the actual cost of a well drilled in 2018 would have cost $3.8
MM more due to the longer laterals and increased use of proppant.
Capex for 2010 Well Parameters, Cost Rates of the Year This represents the application of well
parameters of 2012 with cost rates for the given year. Note that the more simple well design of
2012 would have cost less by 2018 if the current and future indexing was applied.
This illustration helps us see the effect of cost indices and well design changes using 2012 as a baseline.
The gaps shown in Figure 5-19 between 2012 Well Parameters (orange) and 2012 average cost - actual
(green) illustrate the impact of more complex well design on cost. The gap between average cost, actual
(green), and 2012 Cost Rates (red) shows the much higher impact of the declining cost indices.

In conclusion, costs are forecasted to continue to decrease with light recoveries beginning in 2016.
Given that we expect rate decreases within each major cost driver, we can expect little change in the
relative contribution of each (Figure 5-20).

Figure 5-20: Bakken historical and future nominal costs by major cost driver

H. Cost Correlations and Major Cost Drivers


Some relationships between well design and cost are stronger than others. As already mentioned, each
cost component was calculated by measuring the units or amount of a particular well design attribute
and multiplying it by the rate. An analysis of the well design factors contributing to the five primary
cost drivers was conducted for the period of 2010 through 2018. During that time, both the rates and
character for well design attributes changed, rather dramatically in some cases.

When comparing the well design parameter with the cost for that well design parameter over the
specified time period, an R2 value was generated showing the correlation or relative influence as shown
in Figure 5-21. This figure also suggests that for each cost category, there is one well parameter that is
most influential. Fluid costs are guided the most by variance in gel quantities, drilling costs correlate

49
EIA UPSTREAM COST STUDY

highly with drilling efficiency, proppant costs are influenced the most by the cost per lb of proppant and
pumping costs are influenced the most by injection rate. Figure 5-21 also illustrates the relative
importance of each well design parameter as it relates to the total cost of the well.

Figure 5-21: Bakken historical and future nominal costs by major cost driver

Cost per Unit

Depth of well and well


bottom hole pressure
influence drilling costs. As
noted in Figure 5-22, these
have been declining due
primarily to a decrease in
both rig rates since 2012,
which has been accelerated
in 2015, and an increase in
drilling rates per day. We
expect this to level out in the
years ahead as rates stabilize
and drilling efficiency gains
begin to level out. Figure 5-22: Bakken historical and future costs by major cost driver
This same decrease in costs for completion is also evident, although costs per unit of proppant will
continue to drop even after 2015 (figure 5-23). This is likely due to using larger doses of natural

50
EIA UPSTREAM COST STUDY

proppant in lieu of the more expensive artificial proppant. As operators use more frack stages per well,
the economies of scale will also continue to reduce costs here as well.

Figure 5-23: Bakken historical and future nominal costs by major


cost driver
I. Key Take-Aways
Performance concerns: Over time the Bakken has achieved greater efficiencies in well design and
implementation due to the drop of cost rates for the same activities and well design features. Wells
have also become more complex and will continue to do so in the future. However, the Bakken benefits
only marginally from greater production performance per well, as measured by average well EUR.
Design and inputs into Bakken wells will grow. However well performance is likely to lag behind this
since the application of more proppant is not substantially increasing EURs. With the collapse of oil
prices in late 2014, operators have increasingly focused on better site selection. This factor may be
overwhelming any increases in performance due to technological improvement. Going forward waning
prospect quality and in-fill drilling may also contribute to decreased production performance and this
will likely increase unit costs.

Economic decline is diminished by the drop in oil prices. Substantial cost savings will be achieved for the
next several years. This is due to the decreased rates operators have secured from service providers and
not necessarily gains in efficiency. Nevertheless we will continue to see incremental efficiency gains as
operators continue to reduce drill cycle times and drill more wells from single pads.

Influential well design parameters: When modeling well costs in the Bakken the accuracy of some well
attributes may be more important than others when estimating costs. Gel quantities, injection rates,
cost per pound of proppant, and drilling efficiency are the key attributes whose change over time has
greatly influenced costs and caused the most variance.

Decreasing costs: Rates for various materials and services peaked in 2012 when demand for high
horsepower rigs (1000-1500) were in short supply and fracking crews were scarce. As the supply of
these items increased to meet this demand, rates decreased and led to overall cost decreases. This is
despite increases in the amount of proppant and number of stages. This began a general downward

51
EIA UPSTREAM COST STUDY

trend which has accelerated in recent months by as much as 20% due to a very large over supply of
these services.

Operating Costs: There is substantial variability in operating expense. Water disposal, long haul
transport, and artificial lift expenditures are the highest cost items. Given this variability, we would
expect some operators to make substantial improvement. Due to the nature of the services provided,
operating cost reductions will be much less than capital reductions going into 2015. Currently, about
45% of Bakken crude is transported by rail. The difference between long haul transport and pipeline
transport could save an additional $5-$7 per barrel. However, there are no pipelines to either the east
or west coast. Some operators see an advantage to selling into these markets.

52
EIA UPSTREAM COST STUDY

VI. Eagle Ford Play Level Results


A. Introduction and sub-play description
The Eagle Ford is both an oil and gas play located in South Texas Gulf Coast Basin. Since the formation
gently dips (or descends) to the southeast, vertical depths range between approximately 5,000 to
13,000 feet. Oil and volatile oil is found to the northwest, with gas pre-dominating in the deeper
regions to the southeast. Four
sub-plays, each with their own
cost issues, have been identified
and include: Low Energy Oil,
Northeast Core, Western Curve
and Grassy Edge (see Figure 6-
1). Recent activity has been
centered in the oil dominated
Northeast Core and the gas
dominated Western Curve with
over 3,500 wells being
completed in the play during
the past two years (see Figure 6-
2).
Figure 6-1: Location of Eagle Ford and its sub-plays

The play is located proximate to oil markets located in Texas and also has great access to local gas and
NGL infrastructure and markets. Consequently, production of both oil and gas has ramped up quickly to
over 1.5 MM bbls of oil and 6 Bcf of gas per day. Production growth is beginning to taper off, but not as
severely as in the Bakken as operators focus solely on the better performing areas.

Figure 6-2: Historical wells by sub-play

53
EIA UPSTREAM COST STUDY

B. Basic Well Design and Cost (2014)


Total Eagle Ford Cost

Total well costs range from $6.9 MM to $7.6 MM, as shown in Figure 6-3. Drilling costs are lower in the
shallower Low Energy and oil prone Northeast Core sub-plays located to the north and west.
Completion costs are highest in the gas-prone Gassy Edge and Western Curve plays where pumping
rates are highest. However, all areas in the Eagle Ford use similar proppant and fluid amounts.

Figure 6-3: Total Eagle Ford cost by sub-play

Comparison with Published Data

The average Eagle Ford cost of $7.5 MM compares with published costs reported by operators in 2014
as follows:

Operators reported cost from MM$ 5.9, EOG, to $ 9.6 MM, Swift
EP Energy reported $ 7.2 to 7.3 MM
Chesapeake reported $ 6.1 MM
Low Energy Rosetta and EOG reported $ 5.5 to 6 MM
NE Core - Marathon reported cost of $ 7.3 MM
Western Curve - Operators reported cost of $ 5.5 to 7.2 MM
Grassy Edge - Operators reported costs between $ 7 and 7.6 MM

General Well Design Parameters

Table 6-1 below summarizes well design parameters for each sub-play, proppant mixes, amounts, and
horsepower drive cost. We note that the Gassy Edge and Western Curve use the most horsepower.
Casing amounts reflect the variation in total depth and consist of a conductor pipe, and three
intermediate strings. Artificial lift is applied soon after the well comes on stream, but only in oil-prone
Low Energy and NE Core.

54
EIA UPSTREAM COST STUDY

Well Parameter Unit Low Energy NE Core Energy Western Curve Gassy Edge
TVD Ft 8,098 10,857 8,476 9,290
Horizontal Ft 6,264 5,469 5,819 6,655
Formation pressure Psi 4,859 6,514 5,086 5,574
Frac stages # 19 22 20 18
Frac break pressure Psi 6,802 9,120 7,120 7,804
Pumping rate Bpm 57 70 95 96
Horse Power Hp 10,929 17,994 17,994 21,116
Casing, liner, tubing Ft 27,089 34,169 26,592 31,430
Drilling days Days 18 20 18 20
Natural proppant MM lbs 4.93 7.04 5.11 5.02
Artificial proppant MM lbs 2.21 n/a 2.19 1.67
Total Water MM gal 5.89 5.71 6.18 6.85
Total Chemicals Gal 441,793 256,958 294,130 342,575
Total Gel Lbs 58,906 57,102 5,883 6,851
Table 6 1: Properties of typical wells in each sub-play used to calculate costs

Wells in the Low Energy area are drilled to just over 8,000 feet vertical depth and have lateral lengths
averaging nearly 6,300 Ft. Lateral lengths are fairly long, with 19 stages using over 7.1 MM Lbs. of
proppant and 5.9 MM gallons of fluid. Proppant mixes here are fairly high cost with a substantial
component of ceramic sand. Completion fluids are sometimes gel-based, which is typical of oil plays, but
many wells are still completed with slick water, particularly in the gas plays.

Wells in the Northeast Core area are drilled to nearly 10,900 feet vertical depth and have lateral lengths
averaging 5,500 Ft. Lateral lengths are just over standard length using over 7.0 MM Lbs. of proppant
and over 5.7 MM gallons of fluid with 22 frack stages. Proppant mixes here are low cost, using only
natural sand with some of it being 100 mesh. Completion fluids are often gel based, but some slick
water is also used.

Wells in the Western Curve area are drilled to nearly 8,500 feet vertical depth and have lateral lengths
averaging over 5,800 Ft. Proppant and fluid amounts are 7.3 MM Lbs. and 6.2 MM gallons of fluid with
20 frack stages. Proppant mixes here are high cost, consisting of a large portion of ceramics along with
natural sand. Completion fluids are almost always slick water-based.

Wells in the Gassy Edge area are drilled to nearly 9,300 feet vertical depth and have long lateral lengths
averaging over 6,600 Ft. with 18 frack stages. Proppant and fluid amounts are 6.7 MM Lbs. and 7.2 MM
gallons of fluid. Proppant mixes here are fairly high cost, typically using a large portion of ceramics along
with natural sand. Completion fluids are often slick water-based with very few using gel fracks.

C. Operating Costs
Operating costs are highly variable, ranging from $9.00 to $24.50 per boe (see Figure 6-4) and are
influenced by play type, location, well performance and operator efficiency. Overall, these are about $5
to $8 lower than in the Eagle Ford, due primarily to market proximity.

55
EIA UPSTREAM COST STUDY

Figure 6-4: Operating expenses in each Eagle Ford Sub-play


Lease Operating Expense (LOE)

Most of the Eagle Ford Oils leases operating expenses (LOE) are related to artificial lift and maintaining
artificial lift. However the gas plays in the Eagle Ford do not share this cost and are dominated by water
disposal and labor costs. Therefore, LOE costs in the gas plays will be only 60 to 70 percent of those in
the oily portions of the Eagle Ford. Water disposal is a major cost in the Eagle Ford as water production
rates are higher than other plays. The Other category contains common costs, such as pumping,
compression, and other recurring types of costs. The other recurring types of costs are mostly
determined by the cost of energy to run them and are generally negligible. However they make up a
larger share of the total cost for the gas plays (see Figures 6-5a and 6-5b).

Figure 6-5a: Lease operating expense for Eagle Ford Gas wells

56
EIA UPSTREAM COST STUDY

Figure 6-5b: Lease operating expense for Eagle Ford Oil wells

Gathering, Processing and Transport (GPT)


Oil has several market options with substantial pipeline infrastructure, but close access to the gulf coast.
The breakout of the GPT costs is presented below in Table 6-2.

Units Eagle Eagle Eagle Eagle Eagle Eagle


Ford Wet Ford Wet Ford Dry Ford Ford Ford
Gas High Gas Low Gas High Dry Gas Oil Oil
Low High Low
Gas Gathering $/mcf 0.60 0.35 0.80 0.35 0.60 0.35
Gas Processing $/mcf 0.70 0.30 n/a n/a 0.70 0.30
Short Transportation Oil $/bbl 2.50 0.75 n/a n/a 2.50 0.75
Long Transportation Gas $/mcf 0.30 0.20 0.25 0.2 0.30 0.20
Long Transportation Oil $/bbl 3.50 3.00 n/a n/a 3.50 3.00
Long Transportation NGL $/bbl 2.70 2.20 n/a n/a 2.70 2.20
NGL Fractionation $/bbl 2.94 2.52 n/a n/a 2.94 2.52
Water Disposal $/bbl w 3.50 1.00 3.50 1.00 3.50 1.00
Table 6-2: Breakout of GPT costs

Refineries make for low transportation differentials of around $2.00/boe, even when trucking oil and
natural gas liquids. Short haul transportation for oil is the most variable and is determined by proximity
to delivery points.

Eagle Ford gas infrastructure benefits somewhat from prior conventional development, as well as from
close proximity to end markets and ongoing development of new infrastructure. No real issues related
to gas marketing are evident. Some companies benefit from vertical integration and build their own
gathering systems and gas processing plants. NGL fractionation fees are similar to other areas, but fees
for long haul transport of NGLs are low due to close proximity to the Mont Belvieu market.

G&A Costs

General and administrative costs will decrease over time. However this cost is expected to increase
slightly in 2015 for many companies as they have reduced their labor force and are paying severance
compensation.

57
EIA UPSTREAM COST STUDY

Cost changes in 2015

Table 6-3 below summarizes operating cost changes that we expect to see between 2014 and 2015
going forward.

Change 2015
Gas Gathering -3% The operators that operate their own gathering systems will find
that they are saving from lower energy costs, but saving for others
will be marginal.
Gas Processing -3% The operators that operate their own processing plants will find that
they are saving from lower energy costs, but saving for others will be
marginal.
Short Little savings are expected as there were no issues in prior years.
Transportation Oil -3% However as much production is hauled locally by truck, some savings
on fuel costs will be seen. Pipeline costs may not drop much.
Long -3% Lower energy costs will allow for slightly better rates in 2015.
Transportation Gas
Long -3% Those who truck will see saving, but piped oil will not see any
Transportation Oil savings.
Long -3% Better energy cost rates will help lower NGL transportation costs.
Transportation
NGL
NGL Fractionation -5% Fractionation charges have been high but decrease as fuel costs are
low.
Water Disposal +1.80% Many water disposal contracts have fixed rates and some of this will
escalate based on PPI or another indexes. Only companies that
dispose of their own water will see savings.
G&A +5% Severance package/payments due to layoffs are increasing G&A
despite lower future operating cost. Savings will not be realized until
2016.
Artificial Lift -10% Oil field services rates are dropping due to lower activity and lower
input costs rates such as energy.
Artificial Lift -10% Oil field services rates are dropping due to lower activity and lower
Maintenance input costs rates. Maintenance will now be avoided in some cases
where it was profitable at higher prices. Companies that pay a fixed
maintenance may not see better rates in 2015 unless they are able
to renegotiate.
Direct Labor -3% Saving here will be due to fewer operational employees.
Other (pumping, -10% Energy costs savings.
compression, etc.)
Table 6-3: Changes in Eagle Ford operating costs 2014 to 2015

D. Leasing Costs
Lease acquisition costs will depend on if the operator has secured acreage before the play has been de-
risked, as explained in Chapter 1. Figure 6-6 provides recent transaction costs per acre and the
incremental cost to each well that is incurred. Some caution needs to be exercised while interpreting

58
EIA UPSTREAM COST STUDY

this chart as recent transactions are relatively minor. Furthermore many of the exchanges involve
purchase of producing wells, which is not represented in this chart. We note that some operators, such
as Devon and Marathon, have paid handsomely for prime acreage in the Northeast Core oil play, with
per acre charges in the $32,000 to $72,000 range.

We are assuming that each


lateral is going to require 50-80
acres and that two stacked
laterals can be drilled in some of
the areas for a net requirement
of 50-60 acres per well.
Approximately 10-20% of the
acres acquired will not be
utilized. Ultimately we begin to
see that paying $7,000/acre for
50 acres will add up to an
additional $0.4 MM per well.
Figure 6-6: Eagle Ford acreage cost
When we consider the more
extreme cases of paying approximately $50,000/acre in the oil producing sweet spots, we can expect
two to three stacked laterals on a 53-acre area, for approximately 20 -30 acres per well. This still adds
an additional $1.0 MM to $1.5 MM to the cost of each well.

E. Key Cost Drivers and Ranges


Overall, 74% of a typical Eagle Ford wells total cost is comprised of five key cost drivers (see Figure 6-7):

Figure 6-7: Eagle Ford capex breakdown

59
EIA UPSTREAM COST STUDY

Drilling:
o Rig related costs (rig rates and drilling fluids) 16% or $1.2 MM
o Casing and cement 12% or $0.9 MM
Completion:
o Hydraulic fracture pump units and equipment (horsepower) 22% or $1.65 MM
o Completion fluids and flow back disposal 13% or $0.98 MM
o Proppants 13% or $0.98 MM

Range of Costs and Key Drivers

Various cost attributes are classified within each of the five main key drivers as shown in Figure 6-8.
The total cost for each of the five key cost drivers is portrayed with P10/P90 ranges created for each of
the contributing attributes pertaining to such range. These ranges are intended to portray variation
and uncertainty.

Pumping costs, the most costly well


component on average, are
variable with each of the primary
components contributing
substantially to differences in total
well cost. Due to variability found
in the data, formation break
pressures have a range of 5,933 psi
to 10,664 psi, which has the largest
effect on pumping costs. This
creates a range of $ 1.1 MM, with a
potential to increase costs over the
average by $ 0.7 MM and decrease
costs by $ 0.25 MM.
Figure 6-8: Range of cost for attributes underlying key drivers
Drilling penetration rate variability,
from 387 Ft/d to 1,526 Ft/d, creates a drilling cost range of $ 1.0 MM, increasing costs by up to $ 0.7
MM for wells that drill slowly and lowering costs by up to $ 0.3 MM for drilling faster than the average.
Drilling penetration rates are skewed toward faster drilling since is actually quite rare for a well to be
drilled at the slower end of the distribution.

The proppant amount variability, from 3.4 MM Lbs. to 11.6 MM Lbs., creates a proppant cost
distribution of $ 1.8 MM, with the potential to lower costs by just $ 0.7 MM and raise costs by $ 1.1
MM. Most wells will use amounts of proppant on the lower end of the spectrum. However, it is common
for wells to use large amounts too. The fluid cost range for total fluid amount is $ 0.9 MM, potentially
raising costs over the average by $ 0.5 MM and lowering costs by $ 0.4 MM (with fluid amounts ranging
from 3.3 MM gallons to 10.1 MM gallons). The range of vertical depths in the play, from 7,758 Ft. to
11,109 Ft and creates a casing cost range of just $ 0.2 MM. Upward or downward cost movement in this
category is mostly negligible, but is out of the control of the driller.

60
EIA UPSTREAM COST STUDY

F. Evolution of Historical Costs


Historical Well Costs

Between 2008 and 2009: Steel costs rose significantly and created a spike in 2008 that was followed
shortly by a drop due to oil and gas prices weakening in the later part of that year.

Nominal well costs have grown


year-on-year, except for 2013
despite increasing frack intensity
and well dimensions, when fluid
source and disposal options
improved along with completion
service rates (See Figure 6-9). The
rising costs in the Eagle Ford from
2008 to 2012 were a result of
increasing costs rates for
completion, particularly for
Figure 6-9: Historical nominal well cost by major cost driver completion fluids. Since 2012 well
dimensions continued to increase,
but cost rates improved for fluid and frack pumps. Proppant costs have continued to rise, especially
while moving into 2014. This is not solely due to the growth of the amount of proppant used, but also
the mix of proppants have increased in average price from $0.14/Lb. to $0.22/lb. as more completions
relied on artificial proppant. Casing and drilling prices have been fairly constant in recent years with
slight variations due mostly to cost rates and improvements to drilling efficiency.

Between 2006 and 2011, lateral lengths steadily increased until they reached the current length of just
less than 6,000 feet (Figure 6-10). Proppant per well has grown steadily year over year, but feet per
stage has remained constant, which suggests that fluid and proppant concentrations in each stage are
increasing (Figure 6-11). Despite downward pressure on rates from 2013 to 2014, the additional
proppant per well in year 2014 contributed to a slight increase in cost for the well.

Figure 6-10: Lateral length and total depth Figure 6-11: Proppant per well history
history

61
EIA UPSTREAM COST STUDY

Changes in Well and Completion Design

The mix of frack fluids has evolved over the years, beginning with predominately water and slick water
fracks. Then almost immediately in 2011 operators switched to X-link gels with a few still using slick
water. The predominance of X-link gel appears to be a function of drilling more oil wells compared to
gas which typically used slick water
(see Figure 6-12). Well EURs have
increased from 217 kBoe in 2010 to
515 kBoe in 2014. This suggests that X-
link gel fracks and additional proppant
had a positive impact on performance
and that the additional capex was
paying off. Overall play well cost per
Boe has improved from 2010 at
$25.54/Boe to $13.84 in 2014 (see
Table 6-4). Most of the improvements
came during a time when cost rates Figure 6-12: Change in frack fluid use over time
were going down and performance
was increasing dramatically.

Year $/Boe EUR -Boe


2010 25.54 216,958
2011 29.79 227,252
2012 28.08 272,400
2013 21.76 315,541
2014 13.84 514,700
Table 6-4: Vintage drilling and completion unit cost

With lateral lengths holding steady at


6,000 feet, performance has increased
per lateral foot, particularly from 2013 to
2014 (Figure 6-13). This overall increase
in average EUR from 227 kboe in 2011 to
315 kboe in 2013 is likely due to slightly
longer laterals and increases in proppant
(Figure 6-14). At the same time
efficiencies in drilling and completing
have also reduced costs since 2011 (Table
6-4). In 2014, EURs rose dramatically and
we see the trend continuing as operators
are more selective in both their oil and
gas drilling locations due to lower Figure 6-13: EUR per lateral foot
commodity prices. over time

62
EIA UPSTREAM COST STUDY

Some of the performance increase is due


to incremental increases in proppant
usage, as boe per proppant also increased
in 2014. Nevertheless, despite the use of
this technology the performance increases
are much more related to site selection
and overall prospect quality. The sweet
spots have been delineated and operators
will drill the best areas as they attempt to
reduce their costs per boe.

Figure 6-14: EUR per Lb. proppant


over time
G. Future Cost Trend
Cost Indices
Eagle Ford development activity is dropping
sharply with little chance of recovery soon.
Active rigs in the play are currently about
100 and expected to drop into the high-70s
by the end of the year. Because the Eagle
Ford is near to Gulf Coast oil refineries, its
production is able to fetch the WTI price
easily. Services and equipment, such as rigs
and pumping units, may be able to move
into the Permian Basin, but there will be a
surplus there as well. However, this may
relieve some pressure on cost reduction.
Overall, cost rates will decrease from 2014
levels by 22% during 2015, and will drop Figure 6-15: Indices for major cost drivers
another 3% in 2016 (Figure 6-15).

Pumping and drilling costs rates are dropping and are expected to be 25 30% lower by the end of
2015, with another 5% decrease in 2016. Rates will begin to recover in late 2016, but will stay low
through 2018. Proppant costs will drop by 20-25% in 2015, largely due to decreases of 36-40% at the
mine gates. The impact on fluid will be less. Due to a forecasted drop of 20% during 2015 in the price of
steel, tubulars, and other fabricated materials will also cost less.

Changes in Well Design

Despite the challenging environment, operators will continue to lower unit costs ($/Boe). The following
trends are expected to continue:

63
EIA UPSTREAM COST STUDY

Lateral length - Average lateral length has


slowly crept upward during the past four
years and is projected to grow to 6,400 feet
(Figure 6-16). Vertical depths should also
remain fairly constant.
Stages - The average number of stages is
projected to remain the same in 2015, but
should reach 22 by 2018 (Figure 6-17).
Furthermore, although lateral lengths are
projected to change, we can expect that
stage spacing reductions will outpace lateral
lengths.
Drilling efficiencies - These have already
been optimized and any changes here will
Figure 6-16: Historical and forecasted total depth
be small with average drillers achieving
1075 Ft/d by 2018, up from 994 Ft/d in
2014 (Figure 6-16).
Proppant - Proppant amounts will increase
from 1,178 Lbs./Ft in 2014 to 1,215 Lbs./Ft
by the end of this year and will flatten out
until 2018 (Figure 6-18). This is consistent
with other plays. Proppant mix is expected
to be focused more heavily on natural
proppants in order to afford more total
proppant. Average fluid use is expected to
increase proportionately, but at a slower
rate than proppant. Gel and chemicals used
Figure 6-17: Historical and forecasted frack stages
are expected to remain the same going
forward as completion fluids types have
been fixed for some time.
More wells being drilled on single drill pads
As more wells occupy single drill pads we
can expect potential cost savings from
shared facilities and other related items,
such as roads, mud tanks, and water
disposal systems. Of the total well cost,
$1.35 MM is based on sharing costs amongst
four other wells. Table 6-5 illustrates how
future drill pad configurations could save
money. For example, historically there was
one zone, namely the lower Eagle Ford,
which was considered a potential target.
Figure 6-18: Historical and forecasted proppant
Operators are currently completing wells in

64
EIA UPSTREAM COST STUDY

at least one additional zone in the upper Eagle Ford / Austin Chalk, bringing in another potential
zone. Pilot programs have also been completed for tighter spaced wells. Thus the potential
exists for up to 16 wells to be drilled from a single pad, perhaps even more. This could save
potentially $900,000. These savings are not likely to apply throughout the entire play, but are
becoming a common practice in the NE Core area. Other similar areas may emerge as well,
illustrating additional potential savings.

Stacked Distance Wells Cost of items


Horizons between per pad related to pad -
wells 2014
Modeled 1 1320 feet 8 $ 1,350,000 Modeled Cost
Traditional View 1 660 feet 8 $ 1,350,000 Development Cost
Potential upside 2 450 feet 24 $ 450,000 Potential New Cost
Difference 1 1.5 3 $ 900,000 Potential Savings
Table 6-5: Potential savings from additional wells being drilled from a single pad

Future Well Costs

Future changes in overall well and completion costs are quantified in forecasted indices, and are
combined with projections in future well design parameters. Figure 6-19 shows both the effect of well
design and indexing on recent historical costs beginning in 2012 and future well costs through 2018:

Avg. Capex, Actual This captures the average nominal total well cost for each year as it actually
is expected to occur. Note the acceleration of the rate declines from 2014 to 2015 despite more
complex well designs of recent
years which are expected to
continue.
Capex for 2012 Cost Rates, Well
parameters of the year This
captures the application of 2012
cost rates to the average well
design of a given future year. Note
that had we held 2012 rates steady
through the forecast period, the
actual cost of a well drilled in 2018
would have cost over MM$ 3.2
more due to the longer laterals and Figure 6-19: Comparison of actual future costs with
increased use of proppant. forecasted indices
Capex for 2012 Well Parameters,
Cost Rates of the Year This represents the application of well parameters of 2012 with cost
rates for the given year. Note that the more simple well design of 2012 would have cost about
$MM 0.7 less by 2018.

65
EIA UPSTREAM COST STUDY

This illustration helps us see the effect of cost indices and well design changes using 2012 as a baseline.
The gap between 2012 Well Parameters (orange) and 2012 average cost - actual (green) illustrates the
impact of more complex well design on cost.
Whereas the gap between average cost - actual
(green) and 2012 Cost Rates (red) shows the
much higher impact of the declining cost indices.

In conclusion, costs are forecasted to drop in


2015 and are expected to start moving slowly
upward after 2016 (Figure 6-20).

Figure 6-20: Drilling and completion nominal cost


forecast

H. Cost Correlations of Major Cost Drivers


Some relationships between well design and cost are stronger than others. As already mentioned each
cost component was calculated by measuring the units or amount of a particular well design attribute
and multiplying it by the rate. An analysis of the well design factors contributing to the five primary
cost drivers was conducted for the period of 2010 through 2018. During that time both the rates and
character for well design
attributes changed.
When comparing the well
design parameter with the
cost for that well design
parameter over the
specified time period, an
R2 value was generated
showing the correlation or
relative influence as shown
in Figure 6-21. This figure
also suggests that for each
cost category, there is one
well parameter that is
Figure 6-21: Drilling and completion nominal cost forecast most influential. Fluid
costs are guided the most
by variance in gel quantities, which is the most influential of all well design factors. Drilling costs
correlate highly with drilling efficiency, proppant costs are influenced the most by the cost per lb of

66
EIA UPSTREAM COST STUDY

proppant and pumping costs are influenced the most by injection rate. Figure 6-21 also illustrates the
relative importance of each well design parameter as it relates to the total cost of the well.

Cost per Unit

Depth of well and well bottom hole pressure influence drilling costs. As noted in Figure 6-22, these have
been declining due primarily to a decrease in both rig rates since 2012. Due to high rig counts in the
Eagle Ford and demand for rigs, the cost rate increased slightly in 2014, increasing cost per foot and cost
per psi of pressure. Falling rig counts in 2015 have accelerated these cost decreases. We expect this to
level out in the years ahead as rates stabilize and drilling efficiency gains begin to level out.

This same decrease in costs for completion is also evident, although costs per unit of proppant will level
out after 2015 (Figure 6-23). While more proppant per well is likely to increase, the mix of natural and
more expensive artificial proppant is not likely to change. As operators use more frack stages per well,
the economies of scale will also continue to reduce costs through 2015, but afterward this will level out
as more proppant is used.

Figure 6-22: Completion cost rates Figure 6-22: Completion cost rates

I. Key Take-Aways
Performance increases: Over time the Eagle Ford has achieved greater efficiencies in well design and
implementation due to the drop of costs for the same activities and well design features. Proppant use
is increasing. However unlike the Bakken, this increase in proppant usage correlates with increased
production performance. Nevertheless average proppant amounts are nearly double that of the
Bakken. The large increase in 2014 is attributable to both technological improvement and better site
selection. With the collapse of oil prices in late 2014, operators have to continue to increasingly focus
on better site selection. This factor may ultimately supersede any increases in performance due to
technological improvement. As this play matures, declining prospect quality and in-fill drilling may also
contribute to decreased production performance. Ultimately unit costs are likely to level out and rise
within the next 3 to 4 years.
Economic performance was superb in 2014 as prices remained high and performance improved, but has
now become diminished by the drop in oil prices. While substantial cost savings will be achieved for the

67
EIA UPSTREAM COST STUDY

next several years, most of this is due to decreased rates that operators have secured from service
providers and not necessarily due to gains in efficiency. Nevertheless, we will continue to see
incremental efficiency gains as operators continue to reduce drill cycle times and drill more wells from
single pads, with as many as 12-16 wells per pad in some areas.

Influential well design parameters: When modeling well costs in the Eagle Ford, the accuracy of some
well attributes may be more important than others when estimating costs. The key attributes, whose
change over time has most greatly influenced costs and caused the most variance in costs, are gel
quantities, injection rates, cost per pound of proppant and drilling efficiency.

Decreasing costs: Rates for various materials and services peaked in 2012 when demand for high
horsepower rigs (1000-1500) were in short supply and fracking crews were scarce. However, some rate
increases are evident in 2014 due to high rig counts. Ultimately the drastic reduction of over 50% in the
Eagle Ford rig count contributed to a large average drop in costs of 25% . This downward trend is
expected to continue for another year. However, as prices recover and activity picks up, cost increases
are likely to occur at a faster rate than efficiency gains.

Operating Costs: There is substantial variability in operating expense, with water disposal, and artificial
lift expenditures being the highest cost items. Proximity to markets and abundant infrastructure
contribute to lower transport fees. Furthermore, differentials to WTI and HH are very low (less than
5%), making this an attractive location. Due to the nature of the services provided, operating cost
reductions will be much less than capital reductions going into 2016. We can expect most future
decreases to be related to reductions in artificial lift for oil wells and compression for both oil and gas
wells.

68
EIA UPSTREAM COST STUDY

VII. Marcellus Play Level Results


A. Introduction and sub-play description
The Marcellus gas play, located in the mountains of Pennsylvania and West Virginia, includes areas with
wet and dry gas. Five sub-plays were identified based on high performance variations and depths in the
formation. This includes: Liquids
Rich, Southwest Core, Periphery,
Super Core, and Northeast Cost
(see Figure 7-1). Drilling within all
sub-plays has leveled off in the
past three years (Figure 7-2).
Production began in 2007 and has
ramped up quickly to nearly 16
Bcf/day, making it by far the
largest gas play in North America.
Consequently, the Marcellus
serves an over supplied gas
market which precipitated drops
in gas price and increased
pressure to reduce the number of Figure 7-1: Location of the Marcellus and its sub-plays
wells being drilled in the play.

Figure 7-2: Marcellus well spuds

Much of the value derived from the Marcellus is from NGL sales, mainly from the Liquids Rich gas area
where current drilling is most active. NGLs are processed locally and are either shipped to the Gulf
Coast or are marketed locally. Lack of processing and transportation infrastructure is being overcome by
new and projected capacity. Production is expected to continue to grow there significantly and thus
more infrastructure will be needed. The Marcellus benefits from being fairly close to market. However

69
EIA UPSTREAM COST STUDY

logistically infrastructure is still lacking and transport fees are high. Also, water disposal is extremely
expensive, averaging over $5/bbl in some areas.

B. Basic Well Design and Cost (2014)


Total Marcellus Cost
Total well cost ranges from $4.9 MM
to $7.9 MM, as shown in Figure 7-3.
Variation in lateral length and
completion design amongst the plays is
also reflected in highly diversified cost
for drilling and completion. The SW
Core and Super Core are the deepest
plays. Proppant use in the Northeast
Core, a highly prolific area, is about
50% that of the other plays. Hence the
completion costs are much lower.

Comparison with Published Data

The Marcellus has a wide range of Figure 7-3: Total Marcellus cost by sub-play
cost. The average Marcellus cost of
$6.4 MM compares with published costs reported by operators in 2014 as follows:

Operators reported well cost ranging from $ 4.8 MM to $ 7.6 MM, with Range reporting the
lowest and Consol reporting the highest.
Rex, EQT and Talisman reported costs from $ 5.6 MM to $ 5.7 MM.
Chesapeake reported an average cost of $ 7.3 MM.
Marcellus NE Core Corrizo reports 22 stage wells at a cost of $ 6.3 MM.
Marcellus Super Core - Cabot reported costs of around $ 5.8 MM to $ 6.4 MM, depending on
wells per pad, and Chesapeake reported costs of around $ 7 MM.
Marcellus SW Core - Rice reported costs at MM$8.5, with the use of 13 MM Lbs. of proppant.
Marcellus Periphery - Consol reported well costs of $7.6 MM, with the use of SSL technique and
many more stages than the average well.
Marcellus Liquids Rich - Range wells cost $ 4.8 MM, Rex at $ 5.6 MM and EQT at $ 5.7 MM.

General Well Design Parameters

Table 7-1 below summarizes well design parameters for each sub-play. Lateral lengths are longer in the
southwestern areas of the plays than the Super Core and NE Core plays located in north eastern
Pennsylvania. No artificial lift is required.

70
EIA UPSTREAM COST STUDY

Well Parameter NE SW
Unit Super Core Periphery Liquids Rich
Core Core
TVD Ft 7,923 7,520 7,755 7,750 6,425
Horizontal Ft 5,379 5,044 6,550 6,570 6,258
Formation pressure Psi 4,595 4,362 4,498 4,495 3,727
Frack stages # 14 19 29 21 15
Frack break pressure Psi 8,823 8,723 8,996 5,619 5,925
Pumping rate Bpm 86 85 87 89 79
Horse Power Hp 21,387 20,899 22,060 14,095 13,194
Casing, liner, tubing Ft 23,851 22,715 26,680 25,558 22,243
Drilling days Days 17 16 18 18 16
MM
Natural proppant Lbs. 4.45 10.75 8 11.6 9.93
MM
Artificial proppant Lbs. n/a n/a n/a n/a n/a
Total Water MM gal 3.7 8.35 8.45 10.9 8.16
Total Chemicals Gal 240,678 459,269 422,446 490,681 408,037

Total Gel Lbs. n/a n/a n/a n/a n/a


Table 7 1: Properties of typical wells in each sub-play used to calculate costs

Wells in the NE Core are drilled to below 8,000 vertical depth and have lateral lengths averaging
approximately 5,400 feet. The lateral lengths are sufficient for completion with 14 stages, using over
4.45 MM Lbs. of proppant and nearly 3.7 MM gallons of fluid. Note that frack stages for the NE Core play
are less than the other Marcellus plays and proppant usage is significant in all the above listed Marcellus
plays. Proppant mixes are natural and do not contain artificial proppant. Completion fluids are nearly
always water based.

Wells in the Super Core are drilled to 7,520 feet vertical depth and have lateral lengths of over 5,000
feet. These lateral lengths support 19 stages using over 10.75 MM Lbs. of proppant and nearly 8.35 MM
gallons of fluid. Similar to NE Core, proppant mixes are natural and do not contain artificial proppant and
completion fluids are nearly always water based.

Wells in the SW Core are drilled to 7,755 feet vertical depth and have lateral lengths of 6,550 feet. The
lateral lengths are sufficient for completion with 29 stages and 8.45 MM gallons of fluid. Although the
SW Core uses the highest amount of the above listed Marcellus plays, just 8 MM Lbs. of proppant is
used. Similar to other Marcellus plays, proppant mixes are natural and do not contain artificial proppant
and completion fluids are nearly always water-based.

Wells in the Periphery are drilled to 7,750 feet vertical depth and have lateral lengths of 6,570 feet.
These longer lateral lengths are sufficient for 21 stages, using 11.6 MM Lbs. of proppant and 10.9 MM
gal of water. Similar to other Marcellus plays, proppant mixes are natural and do not contain artificial
proppant and completion fluids are nearly always water based.

71
EIA UPSTREAM COST STUDY

Wells in the Liquids Rich are drilled to 6,425 feet vertical depth, the shallowest of the plays, and have
lateral lengths of 6,258 feet. The lateral lengths are sufficient for 15 stages, using 9.93 MM Lbs. of
proppant and 8.16 MM gal of fluid. Similar to other Marcellus plays, proppant mixes are natural and do
not contain artificial proppant and completion fluids are nearly always water based.

C. Operating Costs
Operating costs are highly variable in the Marcellus ranging from $12.36 to $29.60 per boe (Figure 7-4)
and are influenced by play type, location, well performance and operator efficiency. Overall, this play
offers both very high and very low operating costs rates.

Figure 7-4: Total Marcellus cost by sub-play

Lease Operating Expense (LOE)

Most of the Marcellus lease operating expenses (LOE) is related to labor, water disposal, and costs
associated with pumps and compressors. Since the Marcellus does not produce oil, LOE costs are much
lower than in other plays. Water disposal cost rates are high in the Marcellus as most water must be
pushed to Ohio for disposal. However water production is fairly low, making its significance lower than
in other plays. The common costs such as pumping, compression, and other recurring types of costs,
which are mostly determined by the cost of energy to run them, are generally negligible. However, they
make up a larger share of the total cost for the gas plays (see Figures 7-5a and 7-5b).

72
EIA UPSTREAM COST STUDY

Figure 7-5a: Total Marcellus cost by sub- Figure 7-5b: Total Marcellus cost by sub-
play play

Gathering, Processing and Transport (GPT)

Condensate production is handled through battery storage and is picked up by marketers in the field.
Marketers reduce payment by a large differential as production is trucked or railed to Edmonton,
Alberta for use in oil sands processing.

Marcellus Marcellus Marcellus Marcellus


Units Wet Gas Wet Gas Dry Gas Dry Gas
High Low High Low
Gas Gathering $/mcf 0.60 0.50 0.60 0.50
Gas Processing $/mcf 0.60 0.35 n/a n/a
Short Transportation Oil $/bbl n/a n/a n/a n/a
Long Transportation Gas $/mcf 1.40 0.70 1.40 0.70
Long Transportation Oil $/bbl 11.00 8.00 11.00 8.00
Table 7-2: Breakout of GPT costs

Marcellus gas infrastructure is quite substantial, but there is a supply glut in nearby hubs. Reaching the
Gulf Coast markets is more complicated. However there is sufficient capacity to move production south
to fetch better prices than the local differential affords. Gas marketing is based on a series of

73
EIA UPSTREAM COST STUDY

complicated arrangements that potentially allocate production to many different nodes and
destinations. Dry gas in the Marcellus rarely requires processing as its raw production can meet pipeline
specifications. Few companies benefit from vertical integration. Furthermore, gathering and processing
is almost a monopoly as most of the capacity is owned by one company. NGL fractionation fees are
similar to other areas. However, fees for long haul transport of NGLs are very high since production
must be trucked to Mont Belvieu. Ethane production in this play is injected into the gas line maxing out
the thermal content limit for pipelines. Transportation differentials are so high that recovered ethane
often becomes a net cost. There are alternatives for Ethane in this play as Edmonton can receive
production through a specialized Ethane pipeline.

G&A Costs

General and administrative costs will decrease over time. In 2015 this cost is expected to increase
slightly for many companies as they have reduced their labor force and are paying severance
compensation.

Cost changes in 2015

Table 7-3 below summarizes operating cost changes that we expect to see between 2014 and 2015
going forward.

Change 2015
Gas Gathering -2% Most of the saving will be related to energy costs, but contract rates
are sticky.
Gas Processing -2% Most of the savings will be related to energy costs, but contract
rates are sticky.
Short Not applicable.
Transportation Oil n/a
Long Transportation 2% Long haul transportation may go up despite benefitting from energy
Gas cost savings and more companies sending production through the
same pipelines to the Gulf Coast.
Long Transportation -3% There will be some saving for fuel costs.
Oil
Long Transportation -3% There will be some saving for fuel costs.
NGL
NGL Fractionation -2% Many companies are locked into rates by contract, but new rates
may benefit from the current state of shale development.
Water Disposal +1.80% Many water disposal contracts have fixed rates and some of this will
escalate based on PPI or other indexes. Only companies that
dispose of their own water will see savings.
G&A +5% Severance package/payments due to layoffs are increasing G&A
despite lower future operating cost. Savings will not be realized until
2016.
Artificial Lift n/a Not applicable
Artificial Lift -10% Not applicable
Maintenance

74
EIA UPSTREAM COST STUDY

Direct Labor -3% Saving here will be due to fewer operational employees.
Other (pumping, -10% Energy cost and maintenance savings.
compression, etc.)
Table 7-3: Changes in Marcellus operating costs 2014 to 2015

D. Leasing Costs
Lease acquisition costs will depend on whether or not the operator has secured acreage before the play
has been de-risked, as explained in Chapter 1. Figure 7-6 provides recent transaction costs per acre and
the incremental cost to each well that is
incurred. Some caution needs to be exercised
while interpreting this chart as recent
transactions are relatively minor and many of
the exchanges involve purchase of producing
wells, which is not represented in this chart.
We note that some operators, such as Warren
Resources, have paid handsomely for prime
developed acreage with high production at
rates over $66,000 per acre. Figure 7-6: Marcellus acreage cost per well
We are assuming that each lateral is going to
require 80 acres well spacing. Approximately 10-20% of the acres acquired will not be utilized.
Ultimately we begin to see that paying $15,000/acre for 80 acres will add up to an additional $1.3 MM
per well.

When we consider the more extreme cases of paying approximately $20,000/acre in a sweet spot with
access to additional producing zones, we can expect three stacked laterals on a 160-acre area for
approximately 50 -60 acres per well. This adds an additional $1.1MM to $1.3 MM to the cost of each
well.

E. Key Cost Drivers and Ranges


Overall, 75% of a typical Marcellus total cost is comprised of five key cost drivers (see Figure 7-7):

Drilling:
o Rig related costs (rig rates
and drilling fluids) 18% or
$1.15 MM
o Casing and cement 17% or
$1.09 MM
Completion:
o Hydraulic fracture pump
units and equipment
(horsepower) 28% or
$1.83 MM

Figure 7-7: Marcellus capex breakdown

75
EIA UPSTREAM COST STUDY

o Completion fluids and flow back disposal 15% or $0.96 MM


o Proppants 15% or $0.96 MM

Range of Costs and Key Drivers

Various cost attributes are classified within each of the five main key drivers as shown in Figure 7-7.
The total cost for each of the five cost drivers is portrayed with P10/P90 ranges created for each of the
contributing attributes pertaining to such range. These ranges are intended to portray variation and
uncertainty

Pumping costs, the most costly well component on average, is quite variable, with each of the primary
components of pumping cost contributing significantly to differences in total well cost. Due to
variability found in the data, stage numbers have a range of 13 to 40. This has the largest impact on
pumping costs, creating a range of $1.6 MM with potential increase of costs over the average by $ 0.9
MM and potential decrease of costs of $ 0.6 MM.

Drilling penetration rate variability, from 352 Ft/d to 1,193 Ft/d, creates a drilling cost range of $ 0.9
MM. This encompasses a potential increase of costs by up to $ 0.8 MM for wells that drill slowly and
potential decrease of costs of up to $ 0.2 MM for drilling faster than the average. Drilling penetration
rates are skewed toward faster drilling as it is actually quite rare for a well to be drilled at the slower end
of the distribution. However, it does happen occasionally.

The proppant amount


variability, from MM
Lbs. 3.5 to MM Lbs.
12.0, creates a proppant
cost distribution of $ 1.0
MM. This has the
potential to lower costs
by $ 0.5 MM and raise
the costs by $ 0.5 MM.
The fluid cost range for
total fluid amount is $
0.8 MM, raising costs
over the average by $
0.3 MM and lowering
costs by $ 0.4 MM (with
Figure 7-8: Range of cost for attributes underlying key drivers
fluid amounts ranging
from 1.6 MM gallons to 13.6 MM gallons).

Variance in lateral lengths also contributes to the range of fluid, proppant and the number of stages.
The range of lateral lengths in the play is large, from 3,574 Ft to 7,789 Ft, but creates a casing cost range
of just MM$ 0.2. Upward or downward cost movement in this category is mostly negligible, but is well
within the control of the driller and higher costs in this component imply better formation access.

76
EIA UPSTREAM COST STUDY

F. Evolution of Historical Costs


Historical Well Costs

The first wells were drilled in 2006 and were completed in a much simpler model, with very little costs
being applied to completion drivers.

Between 2010 and 2012, nominal well costs


steadily increased from under $5 MM to $7.4
MM. Well costs began to slightly decrease,
remaining around $7.2 MM in 2013 and
decreasing to $6.4 MM in 2014. Although
proppant costs have increased steadily from
2012 to 2014, significant reductions are
apparent in fluid and pumping costs during
that same period since the cost indices for
these items decreased despite increases in
fluid volumes (see Figure 7-9).

Changes in Well and Completion Design Figure 7-9: Historical nominal well cost
by major cost driver
Between 2006 and 2011, lateral length steadily
increased until it began to level out and increase
more slowly to its current length of just less
than 6,000 feet (Figure 7-10). On the other
hand, proppant per well has grown dramatically
year over year and feet per stage has decreased
steadily to its current stage width of 200 feet.
This means that fluid and proppant
concentrations in each stage are increasing
(Figure 7-11). Despite the additional proppant
per well in year 2014, downward pressure on
rates from 2013 to 2014 overcame this Figure 7-10: Lateral length and total depth
proppant cost and costs for 2014 decreased history
somewhat.

The mix of frack fluids has evolved over the years,


beginning with predominately water fracks. In 2008
operators switched to the slick water gels. At the
same time information gathering improved. In
2010 operators began using X-link gels and
increased until 2013. However, it appears that slick
water is becoming the fluid of choice again. Well

77
Figure 7-11: Proppant per well history
EIA UPSTREAM COST STUDY

EURs have increased since 2010. However the cost decreases of 2014 has contributed to a unit cost of
only $5.17 / boe. Please see Figure 7-12.

Figure 7-12: Change in frack fluid type over time

In 2014, EURs finally rose after several years of


no growth despite longer lateral and increased
proppant. With lateral lengths increasing each
year, performance per lateral foot has barely
dwindled (Figure 7-13). The overall increase in
average EUR from 750 kboe in 2010 to 1,100
kboe in 2014 came largely from increased
proppant combined with lateral lengths
extensions (Figure 7-14). It is important to note,
cost improvements are a result of improved cost
rates rather than the improvements efficiencies
in drilling and completions (Table 7-4). Figure 7-13: Change in EUR per lateral foot over
time

Year $/Boe EUR -Boe


2010 7.84 751,684
2011 6.87 1,015,527
2012 7.39 1,007,205
2013 6.27 1,012,928
2014 5.17 1,109,740
Table 7-2: Vintage unit costs and EUR

Figure 7-14: Change in EUR per lb. proppant over time


78
EIA UPSTREAM COST STUDY

G. Future Cost Trends


Cost Indices

Since the Marcellus is a gas play, rig


activity has declined at a slower pace,
drifting from the mid-70s count down
to around 50. This is due to the drop in
gas prices late in 2014. The count is not
expected to drop much further by the
end of 2015. Marcellus has a need for
more infrastructure, and as it is built,
new production immediately takes
advantage of it. This lack of infra-
structure has resulted in a discount of
over a dollar per mcf compared to
Henry Hub. Consequently, activity is
more concentrated in the liquids rich Figure 7-15: Indices for major cost drivers
area. Also as a regional market for
services, equipment, such as rigs and pumping units, will not be able to move easily to other areas. This
may idle service providers and put downward pressure on costs. Overall, costs will decrease from 2014
levels by 14-15% during 2015, with minimal decreases for 2016. See Figure 7-15.

Pumping and drilling costs rates are dropping and are expected to be 15% lower by the end of 2015,
with another 5% decrease in 2016. Rates will begin to recover in late 2016, but will stay low through
2018. Proppant costs will drop by 20% in 2015, largely due to decreases of 35-40% at the mine gates.
The impact on fluid will be less. Due to a forecasted drop of 20% during 2015 in the price of steel,
tubulars, and other fabricated materials will also cost less.

Changes in Well Design

Despite the challenging environment


operators will continue to lower unit
costs ($/Boe). The following trends are
expected to continue:

Lateral length - Average lateral


length has not moved much
during the past four years and
is projected to remain
relatively constant at 6,000
6,200 feet (Figure 7-16).
Figure 7-16: Historical and forecasted total depth
Vertical depths should also
remain fairly constant.

79
EIA UPSTREAM COST STUDY

Stages - The average number of stages


is projected to increase from 32 in
2015 to nearly 38 by 2018 (Figure 7-
18). Also because lateral lengths are
not projected to change, we can
expect that stage spacing will tighten
to a degree.
Drilling efficiencies Continuous
changes here will cause averages
currently at 800 Ft/day to increase as
drillers will achieve over 1,000 Ft/d by
2018. (Figure 7-17).
Proppant - Proppant amounts will Figure 7-17: Historical and forecasted
increase from 1,600 Lbs./Ft in proppant
2014 to 1,700 Lbs./Ft by the end
of this year and will steadily
increase to 2,000 Lbs./Ft by 2018
(Figure 7-18). Superfraccing is the
norm in this play. Proppant mix is
expected to be focused more
heavily on natural proppants in
order to afford more total
proppant. Average fluid use is
expected to increase
proportionately. Figure 7-18: Historical and forecasted stages
More wells being drilled on single
drill pads As more wells occupy single drill pads we can expect potential cost savings from
shared facilities and other related items such as roads, mud tanks, and water disposal systems.
Of the total well cost, $1.23 MM is based on sharing costs amongst eight other wells. Table 7-3
illustrates how future drill pad configurations could save money.

Stacked Distance Wells Cost of items


Horizons between per pad related to pad -
wells 2014
Modeled 1 660 feet 8 $1,230,000 Modeled Cost
Traditional View 1 660 feet 8 $1,230,000 Development Cost
Potential upside 2 660 feet 16 $615,000 Potential New Cost
Difference 1 1 2 $615,000 Potential Savings
Table 7-3: Potential savings from additional wells being drilled from a single pad

80
EIA UPSTREAM COST STUDY

For example, there is currently one stacked zone in the Marcellus which is considered a
potential target. New wells are being completed in the overlying Burkett Shale, which is now
considered a secondary target. This could become a routine objective and thus the potential
exists for up to 16 wells to be drilled from a single pad. This could save potentially $615,000 per
well. These savings are likely to apply in regional markets, mainly in western Pennsylvania, but
not throughout the entire play.

Future Well Costs

Future changes in overall well and completion costs are quantified in forecasted indices, and are
combined with projections in future well design parameters. Figure 7-19 shows both the effect of well
design and indexing on recent historical
costs beginning in 2012 and future well
costs through 2018:

Avg. Capex, Actual This captures


the average nominal total well cost
for each year as it actually is
expected to occur. Note the
acceleration of the rate declined in
2012, despite more complex well
designs of recent years which are
expected to continue
Capex for 2012 Cost Rates, Well
parameters of the year This Figure 7-19: Comparison of actual future costs with
captures the application of 2012 forecasted indices
cost rates to the average well
design of a given future year. Note that had we held 2012 rates steady through the forecast
period, the actual cost of a well drilled in 2018 would have cost $3.7 MM more due to the longer
laterals and increased use of proppant.
Capex for 2012 Well Parameters, Cost Rates of the Year This represents the application of well
parameters of 2012 with cost rates for the given year. Note that the more simple well design of
2012 would have cost less by 2018.

This illustration helps us see the effect of cost indices and well design changes using 2012 as a baseline.
The gap between 2012 Well Parameters (orange) and 2012 average cost - actual (green) illustrates the
impact of more complex well design on cost. Whereas the gap between average cost - actual (green)
and 2012 Cost Rates (red) shows the much higher impact of the declining cost indices.

In conclusion, costs are forecasted to continue to decrease with light recoveries beginning in 2016.
Given that we expect rate decreases in each major cost driver, we can expect little change in the relative
contribution of each (Figure 7-20).

81
EIA UPSTREAM COST STUDY

Figure 7-20 Marcellus historical and future nominal costs by major cost
driver

H. Cost Correlations of Major Cost Drivers


Some relationships between well design and cost are stronger than others. As already mentioned each
cost component was calculated by measuring the units or amount of a particular well design attribute
and multiplying it by the rate. An analysis of the well design factors contributing to the five primary
cost drivers was conducted for the period of 2010 through 2018. During that time both the rates and
character for well design attributes changed, rather dramatically in some cases.
When comparing the well design parameter with the cost for that well design parameter over the
specified time period, an R2 value was generated showing the correlation or relative influence as shown
in Figure 7-21. This
figure also suggests
that for each cost
category, there is one
well parameter that is
most influential. In the
Marcellus, fluid costs
are guided the most by
variance in completion
fluid type, drilling costs
correlate highly with
drilling efficiency,
proppant costs are
influenced the most by
the cost per lb of
proppant and pumping
costs are influenced
Figure 7-21: Marcellus historical and future nominal costs by major cost driver
the most by injection

82
EIA UPSTREAM COST STUDY

rate. Figure 7-21 also illustrates the relative importance of each well design parameter as it relates to
the total cost of the well.

Cost per unit

The well Depth and well formation


break pressure correlate with drilling
costs. As noted in Figure 7-22, these
have been declining due primarily to
a decrease in both rig rates since
2012. This has been accelerated in
2015 and an increase in drilling .We
expect drilling cost per foot to
remain flat in the years ahead as
savings in cost rates will be
overcome by slightly larger well
dimensions.

A similar decrease in costs for


completion is also evident with the
cost per break pressure and cost per Figure 7-22: Drilling cost rates
pound of proppant going down each year (Figure 7-23) for the Marcellus. Cost per formation break
pressure may go up slightly as this may not drive as much of the cost in the future as it once did. As
operators use more frack stages per well, the economies of scale will also continue to reduce the unit
costs here.

Figure 7-23: Completion cost rates

83
EIA UPSTREAM COST STUDY

I. Key Take-Aways
Performance concerns: Over time the Marcellus has achieved greater efficiencies in well design and
implementation due to the decline of cost rates for the same activities and well design features. Wells
have also become more complex and will remain complex in the future, but at a slower pace. With much
of the play de-risked, many areas will continue to be drilled while at lower cost rates. If production
increases continue in following years the cost per boe will continue to fall. This may be hindered by a
resulting drop in the local natural gas price.
Economic performance is diminished by low gas prices. Substantial cost savings will be achieved for the
next several years. Slight efficiency improvements to well design are expected and completions give
additional production potential.

Influential well design parameters: When modeling well costs in the Marcellus, the accuracy of some
well attributes may be more important than others when estimating costs. Drilling efficiency, pounds of
proppant, formation break pressure and lateral length are the key attributes in the Marcellus whose
change over time has greatly influenced costs and caused the most variance in costs. In the Marcellus
the greatest drivers are fluid type, drilling efficiency, the cost per pound of proppant, and slurry injection
rate.

Decreasing costs: Rates for various materials and services peaked in 2012 when demand for high
horsepower rigs (1000-1500) were in short supply and fracking crews were scarce. As the supply of
these items increased to meet this demand, rates decreased and led to overall cost decreases. This is
despite increases in the amount of proppant and number of stages. This began a general downward
trend which has accelerated in recent months by as much as 20% due to a very large over supply of
these services.

Operating Costs: There is limited variability in operating expense, with the greatest ones being water
disposal, long haul transport, and gathering. Given variability is relatively low compared to other plays,
we would expect few operators to make substantial improvements. Due to the nature of the services
provided, operating cost reductions will be much less than capital reductions going into 2015 and will be
much less than will be experienced in other plays.

84
EIA UPSTREAM COST STUDY

VIII. Permian Play Level Results


A. Introduction and sub-play description
The Permian Basin occupies West Texas and
Eastern New Mexico. For decades it was
historically drilled with vertical wells to
access a series of stacked formations. In
recent years four plays have emerged,
namely the Wolfcamp and Bone Spring
horizontal plays located in the Delaware
Basin, and the horizontal Wolfcamp and
vertical Spraberry located in the Midland
Basin (see Figure 8-1).

In this study we have not generally included


the vertical wells when computing averages
and trends. We have grouped the single
Midland Basin play and the two Delaware
basin plays. These plays are located in a
remote arid desert area that suffers from
water sourcing issues, but gas, oil, and Figure 8-1: Location of the Permian Basin sub-
liquids can still be sold locally in Texas. basins
Well costs have grown rapidly since 2012 as the number of vertical wells has fallen off sharply and have
been replaced by horizontal wells with complex completion designs. Oil production is also leveling off as
rigs have dropped from 330 in 2014 to 150 currently. Logistically this play is farther away from markets.
However it is still closer to Cushing and the Gulf Coast than the Bakken. Recent infrastructure additions
have helped offset the high transport fees that in the past hurt profitability in the region.

B. Basic Well Design and Cost (2014)


Total Permian Cost

Total well cost ranges from $6.6 MM to $7.8


MM, excluding Spraberry, as shown in Figure
8-2. Consistency in TVD, lateral length,
pressure, and completion design amongst
the horizontal plays is also reflected in
similar costs amongst the sub-plays cost for
drilling. Completion costs are driven by
lateral lengths that range from 5,000 feet in
the Bone Spring to 7,200 feet in the Midland
Wolfcamp. Proppant use is also much
greater in the Midland Wolfcamp play. Figure 8-2: Permian historical wells

85
EIA UPSTREAM COST STUDY

Comparison with Published Data

The average Permian cost of $7.5 MM


compares with published costs
reported by operators in 2014 as
follows:

Operators reported cost from


MM$ 5.5 to MM$ 12.3, with
Approach reporting the lowest
and Energen reporting the
highest.
Concho, Laredo, EP Energy,
and EOG reported cost from $ Figure 8-3: Total Permian cost by sub-play
6.2 MM to $ 7 MM.
Rosetta reported costs of $ 8.5 MM, but these wells were very deep.
Bone Spring - Concho reported costs of $ 5 MM to $ 7 MM.
Wolfcamp Delaware - Operators reported cost of $ 7 MM to $ 8.5 MM.
Wolfcamp Midland - Operators reported cost of $ 5.5 MM to $ 8.6 MM.
Spraberry Energen and Diamondback reported cost of $ 2.5 MM.

General Well Design Parameters

Table 8-1 below summarizes well design parameters for each sub-play. Proppant mixes, amounts, and
horsepower drive costs. We note that Midland Wolfcamp uses the most proppant and is almost entirely
natural. Casing programs are uniform with a conductor pipe, two strings, and a liner generally used. It is
common practice for artificial lift to be installed soon after the well comes on stream.

Wolfcamp Wolfcamp
Well Parameters Unit Bone Spring Spraberry
Delaware Midland
TVD Ft 9,715 10,644 7,952 8,996
Horizontal Ft 4,967 5,578 7,257 0
Formation pressure Psi 5,829 6,386 4,771 5,398
Frack stages # 12 20 28 8
Frack break pressure Psi 9,326 8,941 7,157 7,557
Pumping rate Bpm 70 59 78 61
Horse Power Hp 18,401 14,869 15,735 12,993
Casing, liner, tubing Ft 29,112 32,807 29,169 22,086
Drilling days Days 25 23 20 11
Natural proppant MM Lbs. 3.07 4.82 8.82 0.83
Artificial proppant MM Lbs. 1.08 1.39 n/a n/a
Total Water MM gal 6.21 6.25 8.74 0.77
Total Chemicals Gal 372,587 312,658 436,836 38,545
Total Gel Lbs. 186,294 187,595 87,367 7,709
Table 81: Properties of typical wells in each sub-play used to calculate costs

86
EIA UPSTREAM COST STUDY

Wells in the Wolfcamp Delaware play are drilled over 10,600 feet vertical depth and have lateral lengths
averaging nearly 5,600 feet. Lateral lengths are moderate, but still support 20 stages with over 6.2 MM
Lbs. of proppant and nearly 6.6 MM gallons of fluid. The proppant mix is fairly high cost, which is
primarily cheap, natural, and mixed with a lot of ceramic sand. Completion fluids are mostly gel-based
with few wells completed with slick water. Surface casing is not reported in this area and it is assumed
that wells use only three casing strings, completed with production tubing. The oil production in this
play benefits from artificial lift.

Wells in the Wolfcamp Midland play are drilled nearly 8,000 feet vertical depth and have lateral lengths
averaging nearly 7,300 feet. Lateral lengths are very long and support 28 stages with over 8.8 MM Lbs.
of proppant and nearly 9.2 MM gallons of fluid. The proppant mix is low cost, which is primarily cheap
natural sand with some 100 mesh. Completion fluids are either gel or slick water based. The wells are
cased with a standard surface casing and three additional strings completed with production tubing. The
oil production in this play benefits from artificial lift.

Wells in the Bone Spring play are drilled over 9,700 feet vertical depth and have lateral lengths
averaging nearly 5,000 feet. The short lateral lengths only support 12 stages with over 4.1 MM Lbs. of
proppant and nearly 6.6 MM gallons of fluid. The proppant mix is high cost with a lot of variation
between wells, which is primarily cheap natural sand with significant amounts of resin coated or ceramic
sand. Completion fluids are either gel or slick water based. The wells are cased with a standard surface
casing and 4 additional strings completed with production tubing. The oil production in this play benefits
from artificial lift.

Wells in the Spraberry play are drilled to 9,000 feet vertical depth on average with any well deviations
adding just a few hundred feet to the wells measured depth. The completion zone is fairly long for a
vertical well which supports 8 stages which use only 0.8 MM Lbs. of proppant and 0.8 MM gallons. The
proppant mix is low cost comprised of only natural sand. Completion fluids are either gel or slick water
based. The wells are cased with a standard surface casing and 3 additional strings completed with
production tubing. The oil production in this play benefits from artificial lift.

C. Operating Costs
Operating costs are highly variable
ranging from $13.32 to $33.78 per boe
(Figure 8-4) and are influenced by
location, well performance, and
operator efficiency. Costs are similar
between the Delaware and Midland
areas, but the Delaware may incur
higher transportation costs due to its
farther distance from markets.

Figure 8-4: Range of operating expenses

87
EIA UPSTREAM COST STUDY

Lease Operating Expense (LOE)

Most of the Permian lease operating expenses (LOE) incurred relate to artificial lift and maintaining
artificial lift. Water disposal costs are significant, but lower than in other plays. The Permian produces
just 0.2 bbl of water for every Boe that is produced. Direct labor and other costs are fairly small relative
to the rest of the costs. However they are similar to other plays. The Other category contains common
costs like pumping, compression, and other recurring types of costs, which are mostly determined by
the cost of energy to run them (Figure 8-5).

Figure 8-5: Breakout of LOE costs


Gathering, Processing and Transport (GPT)

Oil is sent by either pipeline or rail to either Cushing or the Gulf Coast. The range of costs or differential
incurred depends on whether transport is by rail or pipeline. Recently, in 2015, the Permian has
benefitted from additional pipeline capacity that will allow for much less use of rail, and thus bring costs
down dramatically.

Gas has significant options in this play. The Permian is a region that has produced under past
conventional developments and has a great deal of gas infrastructure and access to markets on the Gulf
Coast. Gas plants and gathering systems are often operated by producers, which allows for low GPT
costs in some cases. Current gas processing, fractionation, and transportation rates are in line with
other plays, but can be higher or lower depending on commercial arrangements. See Table 8-2 for the
GPT cost breakout.

88
EIA UPSTREAM COST STUDY

Units Delaware Delaware Midland Midland


High Low High Low
Gas Gathering $/mcf 0.80 0.40 0.6 0.4
Gas Processing $/mcf 1.25 0.25 0.8 0.25
Short Transportation Oil $/bbl. 3.00 0.25 2.5 0.25
Long Transportation Gas $/mcf 0.30 0.20 0.3 0.2
Long Transportation Oil $/bbl. 13.00 4.00 13 4
Long Transportation NGL $/bbl. 9.78 4.13 9.78 3.04
NGL Fractionation $/bbl. 4.00 2.00 3.6 2.25
Water Disposal $/bbl. water 3.00 2.00 4 2.5

Table 8-2: Breakout of GPT costs

G&A Costs
G&A costs range between $2.00/ boe and $4.00/boe. These may increase during 2015 due to layoffs
and severance pay outs, but will be reduced over time due to staff reductions

Cost changes in 2015

Table 8-3 below summarizes operating cost changes that we expect to see between 2014 and 2015
going forward.

Item Change Description of change for 2015


Gas Gathering -3% Current contracts are sticky, but new contracts will benefit from energy
cost savings. Vertically integrated companies will benefit the most.
Gas Processing -3% Current contracts are sticky, but new contracts will benefit from energy
cost savings. Vertically integrated companies will benefit the most.
Short Will benefit from improved fuel cost rates.
Transportation Oil -3%
Short -5% Improved infrastructure will allow for more piping of production, but
Transportation Gas many operators will incur the same costs as 2014.
Long Transportation -60% Less reliance on rail given new pipeline capacity.
Oil
Long Transportation -5% Some improvement to energy costs, but many will incur the same cost
NGL as 2014.
NGL Fractionation 0% Little change expected.
Water Disposal +1.80% Many water disposal contracts have fixed rates. Some of this will
escalate based on PPI or another index. Only companies that dispose of
their own water will see savings.
G&A +5% Severance package/payments due to layoffs are increasing G&A despite
lower future operating cost. Savings will not be realized until 2016

89
EIA UPSTREAM COST STUDY

Item Change Description of change for 2015


Artificial Lift -10% Oil field services rates are dropping due to lower activity and lower
input costs rates like energy.
Artificial Lift -10% Oil field services rates are dropping due to lower activity and lower
Maintenance input costs rates. Maintenance will now be avoided in some cases
where it was profitable at higher prices. Companies that pay a fixed
maintenance may not see better rates in 2015 unless they are able to
renegotiate.
Direct Labor -3% Saving here will be due to fewer operational employees.
Other (pumping, -10% Energy costs savings.
compression, etc.)
Table 8-3 Changes in operating expense going forward

D. Lease Costs
Lease acquisition costs will depend on if the operator has secured acreage before the play has been de-
risked, as explained in Chapter I. Figure8-6 provides recent transaction costs per acre and the
incremental cost to each well that is incurred.

We are assuming that each lateral is going to require 80 acres for Delaware wells and 60 acres in the
Midland per well. Approximately 10-20% of the acres acquired will not be utilized. Ultimately we see
that paying $15,000/acre will add up to an additional $ 1 to 1.3 MM per well. Acreage costs have
increased in recent transactions as the Permian has been identified as a great producer on par with the
Bakken. Furthermore, much of the play has been de-risked for unconventional development.

Figure 8-6: Historical leasing costs

90
EIA UPSTREAM COST STUDY

E. Key Cost Drivers and Ranges


Overall, 74% of a typical Permians total cost, excluding vertical Spraberry areas, is comprised of five key
cost drivers (see Figure 8-7):

Drilling:
o Rig related costs (rig
rates and drilling fluids)
17% or $1.28 MM
o Casing and cement 13%
or $0.98 MM
Completion:
o Hydraulic fracture pump
units and equipment
(horsepower) 26% or
$1.95 MM
Figure 8-7: Permian spending breakdown
o Completion fluids and
flow back disposal 19% or $1.43 MM
o Proppants 17% or $1.28 MM

Range of Costs and Key Drivers

Various cost attributes are


classified within each of the five
main key drivers certain as shown
in Figure 8-8. The total cost for
each of the five cost drivers is
portrayed with P10/P90 ranges
created for each of the
contributing attributes pertaining
to such range. These ranges are
intended to portray variation and
uncertainty

In the Permian the pumping costs,


the most costly well component
on average, is highly variable with
each of the primary components
of pumping cost contributing to
substantial differences in total
well cost. Due to variability found
in the data, stage numbers have a Figure 8-8: Range of cost attributes underlying key
range of 11 to 37 which have the
largest effect on pumping costs. This creates a range of MM$ 2.1, increasing costs over the average by $
1.5 MM and lowering them by $ 0.7 MM.

91
EIA UPSTREAM COST STUDY

Drilling penetration rate variability, from 279 Ft/d to 1,158 Ft/d, creates a drilling cost range of $ 1.3
MM. This incorporates potential cost increases of up to $ 1.0 MM for wells that drill slowly and cost
decreases of up to $ 0.3 MM for drilling faster than the average. Drilling penetration rates are skewed
toward faster drilling since it is actually quite rare for a well to be drilled at the slower end of the
distribution. However this does happen occasionally.

The proppant amount variability, from MM Lbs. 3.0 to MM Lbs. 12.4, creates a proppant cost
distribution of MM$ 1.7. This may potentially lower costs by $ 0.5 MM and raise costs by $ 1.2 MM. The
fluid cost range for total fluid amount is $ 1.3 MM, raising costs over the average by MM$ 0.6 and
lowering it by $ 0.7 MM (with fluid amounts ranging from 2.3 MM gallons to 11.7 MM gallons).

Variance in lateral lengths also contributes to the range of fluid, proppant, and the number of stages
ranging from 4,401 Ft to 8,666 Ft. The range of vertical depths in the play is also large, from 6,688 Ft to
11,147 Ft, but creates casing cost range of just $ 0.2 MM. Upward or downward cost movement in this
category is mostly negligible.

F. Evolution of Historical Costs


Historical Well Costs

Initially, in the Delaware Basin, wells had short lateral sections and small completions. Drilling and
casing make up most of the well cost. Because of larger wells with more stages, nominal well costs in
the Delaware Basin grew year-on-year until
2013 when pumping and frack fluid costs
decreased due to improved completion
service markets. Overall well design and
completion intensity have grown with frack
stages doubling since 2009 and thus
increasing proppant costs. However in
recent years lateral lengths have decreased.
The increase in cost from 2013 to 2014 is
related to increased stages with longer
lateral lengths, higher power pumping, and
increased formation pressures.

Nominal well costs in the Midland area grew


year-on-year until 2013 when water cost
improved so much that total well cost
decreased. This is despite increasing well
dimensions and frack intensity. Overall, well
design and completion intensity has grown
with frack stages doubling since 2009 and
thus increasing proppant costs. However in Figure 8-9: Delaware and Midland historical nominal
recent years lateral lengths have decreased. well cost by major cost driver
Improvements in pumping costs since 2012 are mostly attributable to more supply of frack equipment
and personnel (Figure 8-9).

92
EIA UPSTREAM COST STUDY

Changes in Well and Completion Design

Lateral lengths have increased in both basins. However, in recent years the increase has tapered off.
The Delaware Basin is the exception, where lateral lengths took a big jump in 2014. This coincides with a
large increase of over 2 MM Lbs of proppant in Delaware wells that year. Lateral lengths have always
been large in the Midland Wolfcamp, averaging over 7000 feet, but proppant amounts which were
large when
the play
began have
soared to
over 10 MM
Lbs per well.
This suggests
that
proppant
concentratio
ns are
increasing.

The large
increases in
the Delaware
Basin may
suggest that
operators Figure 8-11: Delaware and Midland -
Figure 8-10: Delaware and Midland -
are Proppant per well history
Lateral length and total depth history
beginning to
use similar
completion techniques
there as well. This will
surely increase costs in
the Delaware Basin.
Despite downward
pressure on rates from
2013 to 2014, this
additional proppant per
well in year 2014 (in the
Delaware Basin)
contributed to a slight
increase in cost for the
well.

The mix of frack fluids


changed between 2009
and 2011 in both basins.
Figure 8-12: Midland and Delaware - Change in frack fluid use over time

93
EIA UPSTREAM COST STUDY

In the Midland Basin, operators switched to X-link gels and slick water. Slick water is becoming more
popular. In the Delaware Basin the more costly Gel and X-link gel are the fluids of choice.

Well EURs have improved in both basins up through 2013. This suggests that the completion programs
in each basin are working. However, the unit costs ($/Boe) are fluctuating. For example in the Delaware
Basin unit costs are increasing despite a large increase in EUR. In the Midland Basin a much higher EUR is
required in 2012 to generate the same unit cost of just over $45 as was generated in 2010. This
illustrates the need to contain and drive down costs.

Year Delaware $/Boe Delaware EUR -Boe Midland $/Boe Midland EUR -Boe
2010 12.76 314,085 55.07 86,134
2011 10.01 476,799 50.23 147,625
2012 10.62 511,043 54.56 171,834
2013 8.92 577,152 39.10 215,921
2014 9.76 641,488 39.77 185,136
Table 8-4.1: Midland Vintage Unit costs and EUR Table 8-4.2: Delaware Vintage Unit costs and EUR

The Midland was a fairly immature play until recently. It has experienced large improvements since 2010
in both well performance and in well economics (Table 8-4.2). Lateral length increases have staggered
over the last couple of years in the Midland while EUR per well dropped (Figure 8-13). This is mostly due
to exploration attempts expanding the play into less tested areas where shorter lateral lengths were
used. Future Midland development will focus on the core areas and increasing lateral lengths in those
areas to maximize production. Cost per boe worsened going into 2014, but this will improve going
forward as less risky well locations are drilled with better well designs.

Figure 8-13: Midland and Delaware EUR and lateral length

The Delaware, another play coming of age, holds a slightly different story where well design growth has
improved the EUR per well with lateral lengths moving from 4,000 feet to over 5,000 feet (Figure 8-13).
Furthermore, proppant jumped over 50% from 2013 to 2014 (Figure 8-14). However, well economics
have not benefitted much. Despite increasing EURs, the cost per Boe has grown nearly a dollar since

94
EIA UPSTREAM COST STUDY

longer laterals are being drilled with greater completion intensity. Under the new cost environment in
2015 it is expected that well design will continue to grow. This will provide even more production per
well and at better economics than in the recent past.

Figure 8-14: Midland and Delaware - Change in frack fluid use over time

G. Future Cost Trends Future Cost Trends


Cost Indices

Permian development activity is dropping sharply with little chance of recovery soon. Active rigs in the
combined Delaware and Midland Basin plays are down to about 150 from a high of 330 in 2014. Before
the oil price decline, infrastructure was not sufficient to transport oil to Gulf Coast or Cushing. There was
also a large differential to WTI penalty of $6 to $12. Recent additions of take-away capacity have
alleviated the bottlenecks and almost
completely erased the differential penalty.
This provides some cushion to the oil price
decrease. Nevertheless, like other
locations there is great pressure on service
providers to reduce costs. Overall, costs in
the Delaware Basin will decrease from
2014 levels by nearly 23% and the Midland
Basin will decrease from 2014 levels by
over 20% during 2015. The Delaware Basin
will not see costs drop further in 2016.
However the Midland Basin will drop
another 1%.

Pumping and drilling costs rates are


Figure 8-15: Indices for major cost drivers of the
dropping and are expected to be 25 30%
Midland and Delaware Basins
lower by the end of 2015, with another 5%
decrease in 2016. Rates will begin to recover in late 2016, but will stay low through 2018. Proppant
costs will drop by 20-25% in 2015, largely due to decreases of 35-40% at the mine gates. The impact on

95
EIA UPSTREAM COST STUDY

fluid will be less. Due to a forecasted drop of 20% during 2015 in the price of steel, tubulars, and other
fabricated materials will also cost less (Figure 8-15).

Changes in Well Design

Despite the challenging environment operators will continue to lower unit costs ($/Boe). The following
trends are expected to continue:

Lateral length In the Midland Basin, average lateral length will increase by about 500 feet to
over 8,000 feet. In the Delaware Basin some increase is also expected (Figure 8-16). Vertical
depths should also remain fairly constant.
Stages - The average number of stages in the Delaware Basin is projected to increase from 16 to
21 in 2015 and grow to 25 by 2018. In the Midland Basin with its longer laterals, stages will
increase to 35 in 2015 and then to 40 by 2018. (Figure 8-17) Because lateral lengths are not
projected to change, we can expect that stage spacing will tighten slightly.

Figure 8-16: Midland and Delaware well dimensions and drilling efficiency

Figure 8-17: Midland and Delaware number of stages and feet per stages

96
EIA UPSTREAM COST STUDY

Drilling efficiencies These have been sporadic and appear to already be optimized in both
basins. Any changes here will be small, with average gains of about 10% in both basins by 2018.
Current rates in the Midland basin approach 800 Ft/day. The rate in the Delaware Basin is about
700 Ft/day (Figure 8-16).
Proppant - Proppant
amounts will increase from
1200 Lbs. /Ft in 2014 to
1400 Lbs. /Ft by 2018 in the
Midland Basin and from
1000 Lbs. /Ft in 2014 to
1200 Lbs./Ft t by 2018 in the
Delaware Basin. This is
already a high average in
the SuperFrack range, so the
increases are expected to
somewhat taper off (Figure
8-18). Proppant mix is
expected to be focused
more heavily on natural
proppants in order to afford
more total proppant,
particularly in the Midland
Basin. There is a mix of slick
water and X-link gel fracks.
Current trends suggest that Figure 8-18: Delaware and Midland - Historical and forecasted
more slick water fracks may
occur in the Delaware Basin and we may see more X-link gel fracks in the Midland Basin. At any
rate we can continue to see a mix of these frack types.
More wells being drilled on single drill pads. As more wells occupy single drill pads we can
expect potential cost savings from shared facilities and other related items such as roads, mud
tanks, and water disposal systems. Of the total well cost, $0.8 MM is based on sharing costs
amongst four other wells in both basins. Table 8-5 illustrates how future drill pad configurations
could save money.
o Midland Basin - We currently project that two of the multiple Wolfcamp zones could be
accessed from a single pad. If we can increase access to an additional zone and double
spacing to 660-foot spacing, the potential exists for up to 24 wells to be drilled from a single
pad. This could potentially save $700,000 per well. These savings are not likely to apply
throughout the play. These savings will be more focused in localized areas. Nevertheless this
illustrates the level of potential savings.
o Delaware Basin - We currently project that there is either a Wolfcamp or Bone Spring zone
which could be accessed from a single pad. If we can increase access to an additional
Wolfcamp zone and a single Bone Spring zone and double spacing to 660-foot spacing, the
potential exists for up to 24 wells to be drilled from a single pad. This could save potentially

97
EIA UPSTREAM COST STUDY

$667,000 per well. These savings are not likely to apply throughout the play, but will be
focused more in localized areas. However, this still illustrates potential savings.

Stacked Distance Wells Cost of items related


Horizons between wells per pad to pad - 2014
Modeled 1 1320 feet 4 $ 800,000 Modeled Cost
Traditional View 2 1320 feet 8 $ 400,000 Development Cost
Potential upside 3 660 feet 24 $ 133,333 Potential Savings
Difference 2 2 4 $ 700,000 Potential Savings
Table 8-5.1: Midland Basin - Potential savings from additional wells being drilled from a single pad

Stacked Distance Wells Cost of items related


Horizons between wells per pad to pad - 2014
Modeled 1 1320 feet 4 $ 800,000 Modeled Cost
Traditional View 1 1320 feet 4 $ 800,000 Development Cost
Potential upside 3 660 feet 24 $ 133,333 Potential New Cost
Difference 2 2 6 $ 666,667 Potential Savings
Table 8-5.2: Delaware Basin - Potential savings from additional wells being drilled from a single pad

Future Well Costs

Future changes in overall well and completion costs are quantified in forecasted indices, and are
combined with
projections in future well
design parameters.
Figure 8-19 shows both
the effect of well design
and indexing on recent
historical costs beginning
in 2012 and future well
costs through 2018:

Avg. Capex, Actual


This captures
the average total
nominal well cost
for each year as it
actually is
expected to
occur. Note the Figure 8-19: Delaware and Midland - Comparison of actual future costs
acceleration of with forecasted indices
the rate declined
in 2012 in the Midland Basin. The decline accelerated in 2014 to 2015 in the Delaware Basin.
This is despite more complex well designs of recent years which are expected to continue

98
EIA UPSTREAM COST STUDY

Capex for 2012 Cost Rates, Well


parameters of the year This
captures the application of 2012
cost rates to the average well
design of a given future year.
Note that had we held 2012 rates
steady through the forecast
period, the actual cost of a well
drilled in 2018 would have cost
$3.2 MM more in the Delaware
Basin and $4.3 MM in the
Midland Basin due to the longer
laterals and increased use of
proppant.
Capex for 2012 Well Parameters,
Cost Rates of the Year This
represents the application of well
parameters of 2012 with cost
rates for the given year. Note
that the more simple well design
of 2012 would have cost less by Figure 8-20: Midland and Delaware - Historical and
2018. future nominal costs by major cost driver

This illustration helps us see the effect of cost indices and well design changes using 2012 as a baseline.
The gap between 2012 Well Parameters (orange) and 2012 average cost - actual (green) illustrates the
impact of more complex well design on cost. The gap between average cost, actual (green) and 2012
Cost Rates (red) shows the much higher impact of the declining cost indices.

In conclusion, costs are forecasted to continue to decrease with light recoveries beginning in 2016.
Given that we expect rate decreases in each major cost driver, we can expect little change in the relative
contribution of each (Figure 8-20).

H. Cost Correlations and Major Cost Drivers


Some relationships between well design and cost are stronger than others. As already mentioned each
cost component was calculated by measuring the units or amount of a particular well design attribute
and multiplying it by the rate. An analysis of the well design factors contributing to the five primary
cost drivers was conducted for the period of 2010 through 2018. During that time both the rates and
character for well design attributes changed, rather dramatically in some cases.

When comparing the well design parameter with the cost for that well design parameter over the
specified time period, an R2 value was generated showing the correlation or relative influence as shown
in Figure 8-21. This Figure also suggests that for each cost category, there is one well parameter that is
most influential. In the Midland area fluid costs are guided the most by variance in fluid amounts used,

99
EIA UPSTREAM COST STUDY

drilling costs correlate highly with lateral length, proppant costs are influenced the most by the cost per
lb of proppant and pumping costs
are influenced the most by
formation break pressure. In the
Delaware area fluid costs are
guided the most by variance in
fluid amounts used, drilling costs
correlate highly with drilling
efficiency, proppant costs are
influenced the most by the
amount of proppant and pumping
costs are influenced the most by
formation break pressure. Figure
8-21 also illustrates the relative
importance of each well design
parameter as it relates to the total
cost of the well.

Cost per unit

Depth of well and well bottom-


hole pressure correlates with
drilling costs. As noted in Figure 8-
22, these have been declining due
primarily to a decrease in both rig
rates since 2012, which has
accelerated in 2015 and an
Figure 8-21: Midland and Delaware Cost and well parameter correlations
increase in drilling penetration
rates. The Delaware play actually worsened in 2014 and was due to expanding drilling to riskier areas.
We expect drilling cost per foot to improve over 2015. However in the years ahead higher cost rates will
outpace any new drilling efficiencies.

Figure 8-22: Midland and Delaware Drilling cost rates

100
EIA UPSTREAM COST STUDY

A similar decrease in costs for completion is also evident with the cost per break pressure and cost per
pound proppant going down each year (Figure 8-23) for Permian. Cost per formation break pressure
may go up slightly as this may not drive as much of the cost in the future as it once did. As operators use
more frack stages per well, the economies of scale will also continue to reduce the unit costs.

Figure 8-23: Midland and Delaware Completion cost rates

I. Key Take-Aways
Performance concerns: Over time the Permian has achieved greater efficiencies in well design and
implementation due to decrease of cost rates for the same activities and well design features. Wells
have also become more complex and will continue to do so in the future. However, the Midland portion
of the Permian has not benefitted as much as the Delaware. It actually performed worse in 2014 than in
some prior years. With the play returning to core areas in the downturn, well performance is expected
to make up for recent reductions as design and inputs into Permian wells grow. Going forward waning
prospect quality and in-fill drilling may also contribute to decreased production performance and this
will likely increase unit costs.
Economic performance is diminished by the drop in oil price. Substantial cost savings will be achieved
for the next several years. This is due to the decreased rates operators have secured from service
providers and not necessarily gains in efficiency. Nevertheless, we will continue to see incremental
efficiency gains as operators continue to reduce drill cycle times and drill more wells from single pads.

Influential well design parameters: When modeling well costs in the Bakken the accuracy of some well
attributes may be more important than others when estimating costs. Drilling efficiency, pounds of
proppant, formation break pressure and lateral length are the key attributes whose change over time
has greatly influenced costs and caused the most variance in the Delaware. In the Midland area the
greatest drivers are pounds of proppant, TVD, formation break pressure and the cost per pound of
proppant.

Decreasing costs: Rates for various materials and services peaked in 2012 when demand for high
horsepower rigs (1000-1500) were in short supply and fracking crews were scarce. As the supply of
these items increased to meet this demand, rates decreased and led to overall cost. This is despite
increases in the amount of proppant and number of stages. This began a general downward trend

101
EIA UPSTREAM COST STUDY

which has accelerated in recent months by as much as 20% due to a very large over supply of these
services.

Operating Costs: There is substantial variability in operating expense, with water disposal, long haul
transport and artificial lift expenditures being the highest cost items. Given this variability, we would
expect some operators to make substantial improvements. Due to the nature of the services provided,
operating cost reductions will be much less than capital reductions going into 2015. Currently, about
45% of Bakken crude is transported by rail. The difference between long haul transport and pipeline
transport could save an additional $8 per barrel and may make a large improvement to well economics
going forward.

102
EIA UPSTREAM COST STUDY

IX. Deepwater Gulf of Mexico


Each deep-water United States Gulf of Mexico (GOM) field discovery has its own set of features
which influences the costs, including field size, water depth, proximity to other fields, reservoir
depth and pressure, hydrocarbon product, and operator preferences. The impact on development
economics is as follows:

Well drilling costs: The cost of drilling wells in deep water is primarily driven by water depths
and well depths. Technical aspects such as subsalt or, high temperature and high pressure
(HTHP) environments can create challenges and drive costs up.
Field development costs: These costs are related to the installation of equipment in a deep
water environment, such as production platform installations and subsea tiebacks.
Platform construction costs: Supplies, transportation, and installation of infrastructure are key
elements affecting development economics.
Pipeline layout costs: These include the set up and installation of hundreds of miles of deep
water pipelines.

A. Description of major plays


The five core plays in the Deepwater GOM in this study are the Plio/Pleistocene, Miocene, Miocene
subsalt, Lower Tertiary, and Jurassic. There is significant overlap among the plays, but the general play
boundaries are outlined in Figure 9-1. The current focus of most material new field exploration is in the
Lower Tertiary, Miocene subsalt,
and Jurassic plays. The Lower
Tertiary to Pleistocene sandstone
turbidities have been historically
the major exploration targets and
still contain exploration potential.
Currently, structural traps hold
most reserves, while purely
stratigraphic traps only stand for
a small fraction of total reserves.

Companies have moved into


these three growth plays as
Figure 9-1: Deep water GOM major plays
technologies have advanced,
allowing for increases in both Deep water major plays
water and drilling depths. Each growth play offers different opportunities based on a companys risk
tolerance, skill set, materiality requirements, and available capital. In a sustained low oil price
environment, the Lower Tertiary and the Jurassic face challenges due to constrained commerciality and
high break-even costs. Companies must control costs, increase efficiencies, and access improved
technologies to further improve the economics in these growth plays.

103
EIA UPSTREAM COST STUDY

Recent drilling activities and permits


Deepwater GOM development drilling was
steady until the 2010 moratorium following
the Macondo incident. Activity quickly
recovered after the moratorium was lifted
in 2011. Figure 9-2 shows drilling surging in
2012 to compensate for drilling and
production declines in the preceding years,
and 2014 marked one of the highest
activity levels in decades.

Figure 9-2: Deep water GOM development drilling


Exploration and appraisal drilling has responded differently than development drilling after Macondo.
Figure 9-3 indicates that the return to
exploration drilling post-Macondo was
more gradual than development
drilling as companies took the time to
assess the new operating
environment. Exploration drilling
post-Macondo (2011-2014) has
averaged 27 wells per year, with the
sharpest drop occurring in the
immediate aftermath of the incident.
Exploration and appraisal drilling has
gradually increased, reaching 47 wells Figure 9-3: Deep water GOM exploration drilling
in 2014, the highest level in over a decade.

Permit submission data from the US


BSEE (Bureau of Safety and
Environmental Enforcement) is an
important leading indicator of
operator near-term future investment Drilling
behavior in the deep water GOM.
Permitting data in Figure 9-4 for 1H-
2015 shows a continued drop in
permit submissions, as operators have
responded to falling oil prices by
cutting capital expenditure. During
this half year, total submitted well
Figure 9-4: Deep water GOM wells permits
permits declined by 24% from 2H-2014
submitted by type
and 34% from 1H-2014. During this same time period, permit resubmissionsessentially revisions to
existing permit requestsremain close to all-time highs, reflecting a larger regulatory burden in the
GOM post-Macondo operating environment.

104
EIA UPSTREAM COST STUDY

Major operators field discoveries

Most of the field


discoveries since 2004
were led by six operators
Shell, BP, ExxonMobil,
Anadarko, Chevron, and
Hess. Figure 9-5 shows the
deep water discoveries
creaming curve by
operator. BP has become
the largest acreage holder
and most dominant
operator over the last
twenty years and has
established a significant
scale advantage in the Figure 9-5: GOM deep water discoveries by operator
basin. BPs current
development activity is focused on large Lower Tertiary play fields. Shells current exploration focus is
the frontier Jurassic play. Anadarkos significant basin presence grew following the acquisition of Kerr-
McGee in 2006. The companys position is extensive, and it is building a basin portfolio of significant
scale by exploring in three of the growth plays. Chevrons focus has been on the Lower Tertiary play,
which provides materiality for the company and is the main focus of its current activity in the basin.

B. Deepwater development concepts


Drilling

There are two major types of drilling\\\ rigs


for water depths of 1,000 feet and
deeper: semisubmersibles and drill ships.
Semisubmersibles (semis) consist of
floating equipment with a working deck
sitting on top of giant pontoons and
hollow columns. Most semis use anchor
mooring systems, although recently more
semis employ computer controlled
dynamic position systems (DP), which drilling
automatically adjusts with wind and Figure 9-6: Average deep water rig build cost
waves by a global positioning system
(GPS) signal received from a satellite. A drillship is a specially built vessel with a drilling derrick to drill
the wells in water depths of up to 12,000 feet, and its position is also maintained by DP. A drillship has
better mobility, but is less stable in rough water. It is often used in drilling exploration wells. Drillship
build costs are slightly higher than semisubmersibles, and thus the day rate is higher as well. The
estimated average build cost since 2005 is $600MM for semis and $650MM for drill ships (Figure 9-7).

105
EIA UPSTREAM COST STUDY

As a deep water field enters the development phase, the development wells sometimes are drilled from
the production platform with drilling modules, which include the hydraulic, electrical, and load capacity
similar to floating rigs, which are positioned on the decks of the production platforms.

Field Development

The two types of field development schemes in deep


water are: standalone development and subsea
development (Figure 9-7). Deep water wells are either
developed through standalone infrastructure, a
floating production platform or subsea systems that
tieback to a production platform. Subsea development
systems are generally less expensive than standalone
infrastructure and are more suitable for smaller fields
with no nearby infrastructure. Since offshore
operations now extend to water depths of 1,500 feet
and deeper, which are beyond practical fixed platform
limits, floating production systems now provide viable
options in the deep water. Currently there are
approximately 50 floating production platforms in deep
water GOM, and most of them reside in 5,000 feet and
shallower water depths. Infrastructure is scarce
beyond 5,000 feet, especially in the Lower Tertiary
area.

Figure 9-7: Deepwater development


schematic
Selecting the right development system involves assessments of water depth, reservoir character,
location, and accessibility to infrastructure. Figure 9-8 shows four major types of floating production
facilities for deep-water fields: tension leg platform (TLP), spar platform, semisubmersible floating
production platform (semi), and floating production storage and offloading system (FPSO).

Tension leg platforms (TLP) or extended tension leg platforms (ETLP) use a combination of pontoons and
columns, are best suited for water depths of 5,000 feet and shallower, and could have either a dry tree
on the platform or wet tree at the sea floor. Spar platforms float from large diameter cylinders,
weighted at the bottom to keep them upright. They can be used in water depths up to 7,500 feet. Like
TLPs, both dry trees and wet trees can be installed. Semisubmersible platforms, by definition, were
borrowed from semi drilling rig concept and consist of semisubmersible hulls with a production facility
on board. Floating platform, storage, and offloading (FPSOs) facilities are large ships made from either
converted tankers or are newly built, moored with rope chain and have no drilling facility. Subsea wells
are tied back to FPSOs. Production is processed, and oil is stored in the FPSO with periodical offloading
and transporting via shuttle tanker. In the GOM, spars have been the most widely used production
system, followed by TLPs and semisubmersible platforms.

106
EIA UPSTREAM COST STUDY

Figure 9-8: Deepwater FPS and subsea system

Subsea production systems are applied in two scenarios: (1) they connect smaller fields to nearby
existing infrastructure; and/or (2) they can be applied to an area where existing infrastructure is scarce,
especially in emerging plays. In a situation where several discoveries are located close to each other,
but not reachable by directional drilling, a combination of subsea systems and central floating
production platforms are applied for joint field development. Anadarkos Lucius field and Chevrons
Jack/St. Malo fields are typical joint subsea system and FPS developments. Subsea systems can range in
complexity from a single satellite well with a flow line linked to a deep water floater to several well
clusters connected by manifold to a floating facility via flowline and flexible riser.

In addition to technical assesments, ultimate development decisions are dominated by project economic
conditions, which sometimes require collaboration and joint effort between operators. The Hub
concept has been adopted by GOM operators to jointly develop a giant central production platform as
a Host to process and handle production from adjacent multiple fields. Independence Hub, located
on Mississippi Canyon Block 920 in a water depth of 8,000 feet, is the result of a team effort of five E&P
companies and one midstream energy company coming together to facilitate the development of
multiple ultra-deepwater natural gas and condensate discoveries.

Recently, in response to a lower commodity price environment, many of the large operators in the deep
water GOM have been revisiting development options and scenarios, with a near-term focus on
leveraging existing production infrastructure to develop discovered resources through lower cost subsea
tieback developments.

107
EIA UPSTREAM COST STUDY

C. Deep water GOM project cost study


IHS selected four projects representing different plays, development concepts, and technical challenges
and performed high level project cost analysis on each. All the projects included came on stream in late
2014 or 2015. Capital costs for these projects did not include seismic, leasehold capital cost, operating
cost, and decommissioning. All projects are modeled using IHS QUE$TOR and cross-referenced with
published cost data and project development descriptions. Costs are based on 3rd quarter 2014 cost
environment adjusted by historical rig rates for exploration, appraisal, and development wells. Figure 9-
9 presents a high level timeline of the projects. This analysis proves that the Miocene is the most cost
competitive play and although the resources discovered in the Lower Tertiary are quite significant, the
Lower Tertiary requires far more capital and takes much longer to develop.

Figure 9-9: Deepwater project overview

Chevron Big Foot Project (Miocene subsalt & TLP platform)


The Big Foot field is located in the Gulf of
Mexico about 225 miles south of New
Orleans in water depths of 5,200 feet
(Figure 9-10). Discovered in 2006, Big Foot
sits in the Walker Ridge area and holds
estimated total recoverable resources in
excess of 200 million oil equivalent barrels.
The reservoir is Miocene subsalt with
average well depths of about 25,000 feet
SSTVD. It is expected for production to
come on-stream in late 2015.

Chevron developed the field using a dry-


Figure 9-10: Big Foot location map
tree floating, drilling and production facility,
Big Foot ETLP (Extended Tension Leg Platform), which features dry trees and top-tensioned risers. It has
full drilling capabilities including workover and sidetrack capability on the topsides and has a production
capacity of 75,000 barrels of oil and 25 million cubic feet of natural gas per day. The ETLP hull was built
in South East Asia, and integration took place in the US. The ETLP features a push-up type tensioner
system, which allows it to withstand the harsh conditions of the area. A model test of the ETLP
indicates that it would be able to withstand a 1,000-year hurricane and loop currents, which can often
delay and damage installations and can be very costly.

\\\
108
EIA UPSTREAM COST STUDY

We modeled the Big Foot development in QUE$TOR based on the development plan published by
Chevron. Figure 9-11 shows the development schematic: 13 wells, including 3 water injectors drilled
from the platform with dry tree on board, ETLP, and two pipelines transporting oil and gas. The D&C
cost is $81 MM per well, significantly lower than other Miocene subsalt wells. In contrast, the platform
cost is far more expensive than other TLPs in the GOM at $2.67 billion (63% of the total $4.3 billion
project cost). The Big Foot oil pipeline is 40 miles long with a 20 diameter and lies in depths of up to
5,900 feet. The gas pipeline is 17 miles long, and total pipeline cost is $258MM (Figure 9-12).

Figure 9-11: Big Foot Project Schematic

\\\

Figure 9-12: Big Foot cost profile

109

drilling
EIA UPSTREAM COST STUDY

Anadarko Lucius Project (Miocene subsalt & Spar platform with subsea system)
The Anadarko operated Lucius oil field is located in the Keathley Canyon Block with a 7,100 feet water
depth, containing approximately 276 MM Boe 2P recoverable reserves in the subsalt Pliocene and
Miocene sands. Lucius produces oil and gas through a truss spar floating production facility. The spar is
605 Ft-long with a 110 Ft diameter, is installed in 7,100 Ft of water and has a capacity of 80,000 BOPD
and 450 MMcfd. Six subsea wells with well depths of approximately 19,000 Ft TVD are tied back to the
Lucius spar platform, making the total project scheme a combination of a production platform and
subsea system (Figure 9-13). Oil produced by the Lucius spar is exported to the South Marsh Island
(SMI) Area Block 205 Platform by an 18 in diameter - 145 mile long pipeline divided into three sections.

Figure 9-13: Lucius location and development schematic

The field's first oil was produced in January 2015, with total development costs at approximately $2.47
billion (Figure 9-14). D&C costs were approximately $103MM per well. The total project contains four
major cost components: 6 subsea wells D&C, truss spar platform, subsea system, and pipelines. The
subsea system includes one subsea cluster hosting 4 wells and two subsea satellite wells, which are all
connected to a flexible riser via subsea manifold, jumper and flow line. An electrical umbilical is
connected to subsea control panels and transmits information about temperature, pressure and subsea
integrity, as well as electrical power to the subsea equipment.

\\\

110
EIA UPSTREAM COST STUDY

Figure 9-14a: Lucius location map


Figure 9-14: Lucius cost profile

Kodiak Project (Miocene subsalt & subsea tieback HP&HT)


Kodiak is located in Mississippi Canyon Blocks 727 and
771 in water depths of 5,000 feet. The reservoir
contains six pay sands in the Miocene subsalt at
approximately 29,000 feet depth in high pressure and
high temperature (HPHT) reservoirs. The exploratory
well encountered over 380 of Middle and Lower
\\\
Miocene hydrocarbon-bearing sands. Two appraisal
wells have\\\been drilled. Development plans for the field
call for smart completions and subsea tieback wells to
the Devils Tower Truss Spar, located 6.5 miles southeast
(Figure 9-15).

The project schematic (Figure 9-16) consists of a two-


well subsea tieback to the Devils Tower truss spar in
Mississippi Canyon Block 773. Ultra-deep well depth
and high pressure-high temperature (HPHT) create drilling
Figure 9-15: Kodiak location map
tremendous technical challenges from drilling to subsea
drilling
tieback and installation. High pressure and high temperature resistance equipment and design inevitably
add 20% to 30% to the total cost. Figure 9-17 indicates that the D&C costs are estimated to be about
$200MM per well. Several unique technical features are highlighted in this project. First, smart
recompletion design makes sleeve changes and commingling multiple sands available with minimal well
intervention and downtime once production is on-stream. Second, HPHT resistant equipment and well
design are carefully calculated and selected to ensure safety and regulation compliance. All drilling and
completion elements, including conductor, casing, tubing, well head equipment, BOP, mud weight,
cement job, as well as frack pack design, are made to fit harsh downhole conditions. The subsea system,
which includes subsea tree, flowline, and riser, also requires special designs in order to handle corrosive
production fluids. The pipeline will be of a bi-metallic construction,\\\
lined with a corrosion-resistant alloy.
In addition, the host platform modification is also required to handle above-normal arrival pressure and

111
EIA UPSTREAM COST STUDY

temperature. This modification includes processing equipment modification, umbilical and control
system and riser tube installation, which adds about $60MM to total project cost.

Figure 9-16: Kodiak development schematic

\\\

Figure 9-17: Kodiak cost profile

drilling

112

\\\
EIA UPSTREAM COST STUDY

Chevron Jack/St. Malo (Lower Tertiary subsalt and semi platform with subsea system)

Chevron-operated Jack / St Malo deep


water project comprises the joint
development of the Jack and St Malo
oilfields, which are situated
approximately 280 miles south of New
Orleans, Louisiana and 25 miles apart,
in water depths of approximately 7,000
feet (Figure 9-18). Reservoir depths are
in the order of 26,500 feet. Total
recoverable resources of the two fields
are estimated at over 500 MMBoe.
First production was announced in
Figure 9-18: Chevron Jack/St. Malo location map December 2014.

Figure 9-19 shows the fields being co-developed with subsea completions flowing back to a single host
floating production unit (semisubmersible) located between the fields. Electric seafloor pumps are used
to assist production to the host. The Jack and St. Malo host facility has an initial capacity of 170,000
Bopd oil and 42.5 MMcfd of
natural gas, with the
capability for future
expansion. The facility is the
largest semi-submersible in
the Gulf of Mexico (based on
displacement) and has been
\\\
designed to operate for at
least 30 years. The hull was
fabricated and constructed in
South Korea, and topside
facilities were fabricated and
constructed in Ingleside,
Texas. The semi platform acts
as a hub for over 20 subsea
wells, which are divided into
drilling one subsea cluster for the
Jack field and four subsea
clusters for St. Malo. Each
cluster is comprised of subsea
wells, manifolds, pumps and
Figure 9-19: Chevron Jack/St. Malo development schematic other equipment on the
seafloor, and is tied back to
the facility. Water injection wells and subsea booster system are also included. Several new
technologies were developed and applied to develop the Jack/St. Malo fields. According to Chevrons

113
EIA UPSTREAM COST STUDY

announcement, its subsea boosting system is


ranked as the industrys largest seafloor
boosting system, increasing power by 10% over
the previous industry maximum and able to
withstand 13,000 Psi of pressure. A single-trip
multi-zone completion design is able to capture
more layers of reservoir in significantly less
time, saving $25MM per well based on rig time
operating costs. A 140-mile, 24-inch oil export
pipeline marks the first large diameter, ultra-
deep water pipeline in the Walker Ridge area of
Lower Tertiary trend. Figure 9-20 shows that of
Figure 9-20: Chevron Jack/St. Malo cost profile
the total $12 billion estimated project cost, 60%
will be spent on drilling and completion of subsea wells (each costing about $240MM per well, which is
a typical well cost for Lower Tertiary HPHT wells). A cost of $1.5 billion is estimated for the
semisubmersible platform. A $2.5 billion subsea system cost is comprised of 4 subsea clusters, 3
flowlines connecting clusters to risers, 2 flexible risers reaching the platform, 6 water injection subsea
manifolds, and one subsea pump. A HPHT resistant subsea pump costs around $300MM.

D. Detail cost components and cost driver analysis


Drilling and Completion Cost
There
\\\ are four major categories of
deep water drilling and completion
cost: (1) installation or rig and related
cost; (2) materials such as casing and
tubing; (3) equipment such as
wellhead equipment (i.e. Christmas
tree); and (4) insurance. Because
deep water drilling requires a floating
drilling rig, (i.e. semisubmersible or
drillship)
drilling to perform the drilling
operation, the day rate could be over
$500,000 during a period when
demand is high. It is not surprising
that the rig and its related cost could Figure 9-21: Drilling and completion cost component
account for 89% of the total D&C cost
(Figure 9-21).

114
EIA UPSTREAM COST STUDY

Detailed components of the rig and related


costs show that almost 43% are associated with
floating rigs and over 33% are for support and
supply vessels. The day rate and time spent
onsite are key drivers to the total drilling and
completion cost. Figure 9-22 shows total rig
day rates vs. water depth and well depth. The
water depth primarily drives the day rate as
floating drilling rigs are chartered and priced
based on water depth. In addition to the
floating rig, support and supply vessels play an Figure 9-22: Rig & related cost vs. water depth & well depth

important role by providing supplies to


drilling operations. Helicopter and other
services such as logging, cementing, and
testing also are vital to the operation and
could be costly (Figure 9-23). Please note
that special logging service and testing are
optional for offshore development wells,
but are necessary for exploration and
appraisal wells in order to evaluate the
\\\
reservoirs.

While day rates are driven by water depth,


the rig onsite service days are a factor of
well depth and are often goverened by the
geological and technical complexity of the
reservoir. Figure 9-24 shows the correlation
Figure 9-23: Installation - rig & related cost between rig days and well depths under
component
regular reservoir conditions. Under
technically challenging conditions, like drilling
subsalt, HPHT or overbalance/underbalance
reservoirs, it will take much longer
(sometimes over a year) to reach total depth
of the well and may periodically require a
sidetrack if tools are damaged or lost in
borehole. Other factors unique to the deep
water GOM environment, such as hurricanes
and loop currents, can also significantly delay
the drilling operation.
\\\

Figure 9-24: Average rig days by play by operator

115
EIA UPSTREAM COST STUDY

Nevertheless, the combination of day rate and rig service days are unquestionably the key drivers of
total drilling and completion cost. Operators work very hard to secure the rig at the best rate possible
and are motivated to
reduce downtime to
a minimum level.
Offshore
transportation is also
critical, given the
distance from shore
base. Helicopter and
boat expenses, if not
well-managed, could
also contribute to
cost overruns.
Wellhead equipment,
as part of tangible
cost, plays an
Figure 9-25: GOM deep water D&C cost range
important role in the
cost as well. Christmas trees can be installed either at the seafloor well head or on the production
platform, serving as the dry tree. Like onshore wells artificial lifts, such as an electric submersible pump
(ESP), are also commonly applied to the oil well perforation point and could cost between $3MM to
$5MM. Figure 9-25 provides a glimpse of cost ranges for major components of deep water GOM. The
rig cost could swing from $25MM to over $100MM depending on the water depth and well depth, as
could the support and supply vessel cost. Cost for production and wellhead equipment, including ESP,
ranges from $11MM to $15MM. Downhole hardware and the cost of the equipment like conductors,
casing, tubing, and production liner ranges from $7MM to $13MM. Cementing and logging service costs
are between $2MM to $7MM. In a nutshell, the overall drilling and completion costs at normal
reservoir and well conditions are estimated between $60MM to $240MM for wells in water depths from
\\\
7,500 feet to 15,000 feet. The special well design expense for HPHT environments cannot be overlooked
when estimating the cost and can add 20%-30% to the total cost.

Deepwater GOMs range of D&C cost


sensitivity, shown in Figure 9-26,
once more confirms rig costs can
increase as much as 100% over the
average cost as a direct result of rig
rate and rig days. In other words,
drilling offshore deep water cost can be
extremely time sensitive. Major
operators rig days could run from
150 days to almost 300 days
depending on the play. Jurassic play
Figure 9-26: GOM deep water D&C cost sensitivity

116
EIA UPSTREAM COST STUDY

drilling proved to be the most time consuming due to its water depth.

Rig rate is driven by supply and


demand in the short term. Rig
build cost has remained Avg. earned
Earned Rate day rate
New Rate
unchanged over the last 10 years Avg. new fixture day rate
and thus has a minimal impact on
the day rate. Figure 9-27 indicates
that over the last decade, the
\\\
biggest rig rate drop was seen
between 2010 and 2011,
associated with the decreased
activity following the moratorium \\\
after Macondo. While there is still
significant drilling activity taking Figure 9-27: Earned rates vs. fixed rate
place in the deep water GOM, the
short term outlook may be less encouraging. As of thedrilling
first quarter of 2015, average new fixtures rates
(the new contract rate) were at $378,708/d versus $436,482/d for earned rates (existing contract rate)
combining semisubmersible and drillship, reflecting a 13% reduction. Earned rates represent those
contracts signed a year or two ago, while fixed rates are new contract rates, representing the current
market condition. Without a turnaround in new fixture day rates, this would indicate that average day
rates have started declining. The number of operators drilling
looking to secure rig time in 2015 has also
dropped considerably, which reflects the operators concerns of a longer than expected price recovery.
In addition, with the falling of average lead time, operators are confident that they will be able to secure
rigs when needed and that new fixture rates are more likely to fall.

In conclusion, water depth, well


depth, reservoir quality and
productivity are key drivers to
drilling and completion cost. Of the
three major plays, both water depth
and well depth in the Miocene area
are shallower and therefore, the
Miocene has an advantage over
other plays due to its higher
estimated well productivity and
relatively shallower reservoir
depths (20,000 to 24,000 SSTVD).
Most of the drilling and completion
costs for Miocene wells falls
Figure 9-28: Well cost by drilling depth
between $70MM to $165MM
and water depth
(Figure 9-28); however, Miocene subsalt costs could be much higher given the geological complexity and
unpredictability of the play. The Lower Tertiary has experienced the most technical challenges and thus
higher well costs because of the plays lower permeability, deeper reservoirs (>30,000 Ft) and HPHT

117
EIA UPSTREAM COST STUDY

environment. Lower Tertiary subsalt well costs ranges between $150 to $220MM (Figure 9-28). The
Jurassic is located in the deepest water depth which results in the highest well costs at about $230MM.
This estimated well cost assumes a vertical well, wet tree, normal reservoir conditions with downhole
electronic submersible pump (ESP), and no acid gas. If extreme well conditions are considered, such as
high pressure and high temperature or acid gas and heavy oil environment, the well cost could increase
by an additional 20 30 percent.

Field Development Concept Cost Comparison and Floating Production Platform (FPS)
Of the 130+ deep water GOM fields discovered since 2004, there are approximately 60 fields either in
production, under development, or have a sanctioned and selected development plan. Defining and
planning development strategy in the early phase of a project is vital to the success of projects. The
development
concept is
primarily driven
by reserve size,
water depth,
and
infrastructure
availability or
proximity. In
general, the
subsea tieback
is suitable for
smaller fields if
there is a
platform
Figure 9-29: GOM deep water selected projects cost range by pay and field
nearby to tie-in
reserves
to. Most of the
time floating production platforms are needed because of either: (1) larger discovered reserves, and/or
(2) no nearby infrastructure. Figure 9-29 shows the estimated total project costs for the selected 60
fields discovered since 2004 at different development concepts for different plays; these indicate the
correlation between project costs, reserve size (2P) and development concept within the various plays.
The subsea tieback is selected for most of the Miocene fields, with a cost range between $100MM and
$1.5 billion. For associated development wells, spar and subsea tieback project costs range from
$500MM to $6.3 billion, TLP and subsea project costs range from $3 billion to $7.2 billion, and semi and
subsea projects costs range from $100MM to $18 billion. The most expensive projects are all located in
the Jurassic play and are due to water depth and technical challenges. There is only one FPSO
development in the deep water GOM, the Cascade and Chinook project operated by Petrobras, and one
FPSO is under construction, which will be deployed to Stone field operated by Shell. Over the last ten
years, operators in the GOM realized the importance of access to infrastructure and collaboration with
each other to fully utilize the existing or upcoming infrastructure. As a result, the hub concept, which is
several fields jointly developed with a center floating production infrastructure to process hydrocarbon
product from tie-in fields, has been introduced and gradually adapted by major operators. The Perdido

118
EIA UPSTREAM COST STUDY

project, online in 2010, was the first Lower Tertiary hub brought on stream, and was followed by
Cascade/Chinook in 2012 and Jack/St. Malo in 2014. These hubs, with the addition of the Miocene
Subsalt Lucius hub (on stream in early 2015), could spur further Lower Tertiary development, including a
number of unsanctioned Lower Tertiary discoveries that currently appear to be stalled.
Since 2004, there have been
approximately 35 floating production
platform systems (FPS) which have been
built and deployed in the deep water
GOM, and about 50 total deep water
production infrastructures. From the
1990s onward, the overall trend of
platform design has been based on
deeper water depth and larger capacity
(Figure 9-30).

Figure 9-30: GOM deep water production system by water Water depth, capacity, hull design, and
depth topside design including processing
equipment and utility modules drive the floaters cost. TLPs are mostly deployed in water depths of
5,000 feet and shallower. Spars are used in water depths from 2,000 feet to as deep as 8,000 feet. Semis
are mainly deployed in water depths of 5,000 feet and deeper. Drilling facility installation also largely
impacts cost. While a large number of the hulls have been built in shipyards overseas, primarily in South
Korea, Singapore, and Finland, almost all topsides are still built in the US in order to maintain the
integrity and complexity of the technology.

Figure 9-31: Tension leg platform costs by capacity and water depth

119
EIA UPSTREAM COST STUDY

TLPs are more vulnerable to winds and


loop currents and thus are less favorable in
the deep water GOM compared to the spar
and semi. Consequently only six TLPs have
been built since 2003, most costing
between $200MM to $550MM (figure 9-
31). However, the one outlier is Chevrons
recently built Big Foot extended TLP
(ETLP), featuring a dry tree and on-board
accommodations with a large number of
living quarters (Figure 9-31), with an
Figure 9-32: Spar platform costs by capacity and water depth
estimated cost as high as $2.6 billion.

The cost of Spar platforms varies in a relatively


narrower range from $300MM to $800MM.
Perdido is the exception, located in a water
depth of 8,000 feet at an estimated cost of $2.5
billion with one of the largest capacities at
133Mboe/d (Figure 9-32). The capacity of spar
platforms is generally larger than a TLP, and
several TLPs have been designed based on the
hub concept with larger capacities for future tie-
in opportunities. For example, the recently
deployed Anadarko Lucius spar has the highest
capacity of 155 MBoe/d, presumably large
enough to receive the future production from
the Marcus and Spartacus fields. Figure 9-33: Semi-submersible costs by capacity and water depth

Semi platforms consist of a semisubmersible hull


with a production facility on board and most often
they also accommodate a drilling facility. Since
2004, the average newly-built semi costs about
$600MM. The Jack/St. Malo platform, the most
recent in service, was ranked the most expensive
production facility in the GOM with a cost of
$1,550 MM (Figure 9-33). It was designed as a hub
to process production from multiple HPHT
reservoirs in the Lower Tertiary subsalt play. Semis
also have overall larger capacities when compared
to TLPs and spars. Semis are generally used for
larger fields. The average semi capacity built since
2003 is 145 MBoe/d, which is significantly higher

Figure 9-34: FPS hull cost component

120
EIA UPSTREAM COST STUDY

than the average 84 MBoe/d of the TLP and 91 MBoe/d of the spar.
Regardless of the platform type, all floating production systems vary in size and shape. Their primary
difference is the structure that holds them up: the buoyance or hull. FPSs have four common elements:
hull, topsides,
mooring, and risers.
The three major
cost components
for the hull include
fabrication,
materials and
installation (Figure
9-34). The majority
of costs related to
material and
fabrication is steel
purchase and
cutting. While steel
Figure 9-35: FPS topside cost component cost is priced at
$/ton, fabrication is
based on man hours needed. This explains why most of the hulls are fabricated and constructed in South
Korea, China, and Singapore, where labor is less expensive. Nevertheless, the primary driver of hull cost
is the cost of steel.

Similarly, platform topsides also have three major cost components: equipment, installation, and
fabrication (Figure 9-35), in which equipment plays the most important part. Platform equipment is
comprised of oil and gas handling and process equipment, a gas compression facility, water handling,
and power generation/distribution. Most spars and TLPs can accommodate a drilling facility, which adds
30% - 50% incremental cost, depending on the power of the drilling unit (Figure 9-36).

The three main cost drivers for floating production platforms are design, water depth, and topside
weight and capacity. Spar designs are inherently stable due to their deep draft hulls. In addition, they
tend to be much cheaper compared to TLPs
and semis for water depths of 3,000 feet and
deeper. For this reason, they are the most
popular platform in the deep water GOM.
Spars have three buoy systems consisting of
truss, cell, and caisson. Truss and cell costs
are similar. Caisson costs are 20% more
because of the water depth it can withhold.
The floating production system installed at
Perdido field operated by Shell is the worlds
deepest production caisson spar, standing in
8,000 feet water depth. It is also the most
Figure 9-36: FPS cost change on adding drilling unit

121
EIA UPSTREAM COST STUDY

expensive spar in the GOM with an estimated cost of $2.5 billion (Figure 9-32).

Most TLPs and spars can accommodate a


drilling facility, with the rig type from a
tender support vessel (TSV) to workover
rig. The extra weight added to the
topside could be from 1,500 to 2,600
tons and power can be self- contained or
integrated. It costs more to add a drilling
facility on spar than to a TLP because of
the hull design (Figure 9-36).

Figure 9-37 shows the cost changes in


relation to water depth and the number
of FPSs actually deployed in the GOM (by
water depth and type). Due to design Figure 9-37: FPS cost change vs. water depth
limitations, TLPs can only withstand
water depths of up to 6,000 feet. Semis are more costly because a semi vessel has to be purchased and
modified first, and is less sensitive to
water depth compared to a spar. Topside
weight is primarily driven by capacity and
the drilling facility. In the GOM, most of
the TLPs are installed in about 3,000 feet
water depth, and 40% of spars are
concentrated in water depths between
4,500 feet to 5,500 feet. Semis are
\\\ primarily used in water depths over 5,500
feet.

The production capacity is designed based


on reserve size and productivity from the
tie-in fields. Figure 9-38 indicates that in
Figure 9-38: FPS cost change vs. oil capacity the range of 30,000 bbl/d to 200,000
bbl/d, the cost can increase 39% for spars,
34% for TLPs, and 24% for semis. The highest capacity deployed in deep water GOM by FPS type are
BPs Thunder Horse Semi (250,000 bbl/d), Chevrons Tahiti spar (125,000 bbl/d), and Shells Ursa TLP
drilling
(150,000 bbl/d).

\\\

122
EIA UPSTREAM COST STUDY

The cost sensitivity chart (Figure 9-39) shows


that the overall ranking of the three major cost
drivers for floating production platforms:
drilling facility, processing capacity, and water
depth. In addition, other factors, such as the
location of shipyard, installation preference,
insurance and project management, also can
play important parts in terms of cost control.

Figure 9-39: FPS cost sensitivity

Hurricanes and loop currents often cause installation


delays and facility damage, inevitably adding extra
cost. For example, Chevrons Big Foot TLP was
severely damaged recently by a loop current while
preparing for offshore hookup. Chevron estimates it
will
\\\take two years to repair, thus causing significant
delay to production commencement.

Figure 9-40: Subsea system cost component


Subsea sea systems

The deep water and ultra-deep water


discoveries since 2000 significantly
drilling increase the number of subsea tieback
fields. There are three major cost
\\\ components for subsea systems
(Figure 9-40): (1) materials, including
flow line, umbilical and risers; (2)
equipment, including manifold and
jumper; and (3) installation. Subsea
installation often requires ROVs
(remote operated vehicles) to perform
the operation. The umbilical, a
hydraulic powered cord transferring
drilling power, chemicals and communications
Figure 9-41: Subsea system cost change vs. tieback distance
to and from the subsea development,
is literally the lifeline to the subsea system and one of the most expensive pieces of subsea equipment.

123
\\\
EIA UPSTREAM COST STUDY

The primary driver of subsea system cost is tieback


distance to a platform, where cost increases steadily
with distance. Although water depth has some impact,
it is relatively small compared to tieback distance. The
average subsea tieback length in the deep water GOM is
15 miles, and the longest tieback field is Shells
McMensa, consisting of a 68- mile tieback to a fixed
platform developed in 1997. the total cost for two
satellite wells under normal conditions, assuming there
is no gas lift, water injection, chemical treatment nor
acid gas removal, could range from near MM$ 200 to
over MM$ 500 for a 5 mile to 65 mile tie-in distance
(Figure 9-41). Other factors, such as development type
(e.g., satellite or cluster) and whether a subsea booster
system is installed, will have an impact on the cost as
well. Chevrons Jack/St. Malo field, one of the most
expensive tieback projects, includes four subsea clusters
controlling 20 subsea wells and a subsea boosting
system to enhance recovery.

Figure 9-42: Subsea cluster system A single well subsea tieback is designed as a satellite well
with a flow line directly connected to a riser base or
manifold. Multiple well clusters are designed as clusters with multiple subsea distribution units and
umbilical termination assemblies connecting the production wells via connecting manifold to a flowline.
The flowline then reaches to the riser base of the hosting platform, finally arriving at the topside facility
through a flexible riser. Figure 9-
42 illustrates the Jack field subsea
system schematic with one four-
\\\ well cluster and a 9 mile flowline
tie-in to the Jack/St. Malo
semisubmersible floating
production facility. The subsea
cluster system components consist
of commingling and riser base
manifolds, production, test,
injection, gas lift flowlines, a
flexible riser system, umbilical, and
drilling platform controls.

Figure 9-43 compares the cost of


Figure 9-43: Subsea system cost feature single well to different types of subsea systems,
multiple well clusters from single satellite well to
multiple well clusters. They all

124

\\\
EIA UPSTREAM COST STUDY

start with the base design under normal technical conditions and assume a 15 mile tieback to a host
platform and 5,000 feet water depth. Test service is also included. The incremental costs are added
based on certain technical features: (1) High pressure and high temperature will add around 10% to
cost as a special design is required to protect the downstream production or test service from
overpressure, and (2) Chemical injection typically operates through an injection flowline (methanol
injection) into a production well for hydrate suppression. The chemical injection and acid gas removal
are determined from reservoir fluid characteristic and can be very costly, incurring an additional 30% -
45% in costs.

Pipelines
Once oil and gas are separated and processed through the platform, they move through an export riser
to a subsea pipeline and then
either tie-in to an existing
platform or are transported
directly onshore. The diameter
of pipelines is primarily
determined by pressure and
flow capacity. Pipelines in deep
water generally range from 12
to 30 inches in diameter. The
freezing cold environment in
deep water can cause the
following: (1) hydrates to form
in a gas line and plug the

Figure 9-44: Pipeline cost component pipeline, or (2) for oil pipelines,
paraffin, waxy hydrocarbons to
plate the walls of an oil line. To solve these issues, most pipelines are coated with an insulating material
to keep the fluid warm. Often the dehydrating treatment (i.e., methanol injection) is operated from a
topside treating facility and injected into a pipeline in order to remove the hydrate and water vapours.
Oil pipelines are periodically cleaned to remove wax or paraffin build-up in the pipe walls.

The two major components of pipeline costs are materials and installation (Figure 9-44). Materials
consist of mainly line pipe and
\\\
coating. Although most of the
pipelines are made from carbon
steel, other types of material such as
clad 316 stainless, duplex, clad 825
alloy, and CRA also could be applied
in extreme harsh environments and
high capacity pipelines.

The installation costs (Figure 9-45)


drilling are calculated based on the pipe lay
Figure 9-45: pipeline installation cost

125
EIA UPSTREAM COST STUDY

spreads required to install the specific pipeline. They include a lump sum for the shore approach if
needed. Each of the five pipe lay spread vessels (Reel-lay, S-lay without dynamic positioning (DP), S-lay
with DP, J-lay and Solitaire) has a line item for the total time to lay the pipe and mobilize / demobilize
the pipe lay vessel. The number of days required for each vessel is picked up from the installation
durations form. The unit rate cost for each class of vessel includes labor, fuel, consumables and vessel
support systems.

The driving support vessel (DSV) unit cost includes support services, labor, waiting on weather and
consumables and is derived from the installation durations form. The duration shown in the cost sheet is
the sum of the DSV installation and vessel mobilization / demobilization days.

Testing and commissioning equipment


is required on the DSV during testing
and commissioning. The testing and
commissioning duration is dependent
on the pipeline diameter and length.
Additional time is allowed for waiting
and preparation as well as mobilization
/ demobilization of the equipment into
the field. A trench vessel is required
when either a portion or the entire
pipeline is buried. The trenching
duration is dependent on the buried
Figure 9-46: Pipeline costs vs. water depth and size length of the pipeline and whether
there is a shore approach. The duration
shown in the cost sheet is the sum of the trench activity and vessel mobilization / demobilization days.

The four main drivers for pipeline costs are water depth, length, diameter, and capacity. The typical oil
pipeline technical conditions in the deep water GOM are 3,670 feet water depth, 90 mile long, 12 in
diameter, and 46 Mbbl/day capacity. All four cost drivers are interdependent. For example, a larger
pipeline size is required for deeper water depths (>7,000 Ft) and longer distances. Capacity
requirements
\\\ also impacts pipeline size.

Figure 9-46 indicates that there is a


minor cost increase for water depths of
1,000 feet to 6,500 feet. However,
once the water depth is greater than
7,000 feet, the cost could increase by
over 50% and a larger diameter pipeline
is required to sustain the high pressure
environment.
drilling
On the other hand, Figure 9-47 shows a
direct linear correlation between
pipeline length, diameter, and cost. For Figure 9-47: Pipeline costs vs. length and diameter

126
EIA UPSTREAM COST STUDY

distances less than 50 miles, only a 10 inch pipeline is needed and the cost is less than $100MM. For
distances between 50 to 100 miles, a 12 inch pipeline is required, and the cost reaches to $100MM to
$150MM and for distances between 120 to 170 miles, at least a 14 inch pipeline is needed and cost
jumps to $200MM to $250MM. Lastly, when the distance is 170 miles, at least a 16 inch pipeline is
required, and the cost could reach as high as $300MM.

Figure 9-48 demonstrates how the costs


change along with the capacity and size.
The Big Foot project export pipeline, a
140 mile, 24 inch oil export pipeline
marks the first large diameter, ultra-
deep water pipeline in the Walker Ridge
area of Lower Tertiary trend, with an
estimated cost of $800MM, inclusive of
a gas pipeline.

Figure 9-48: Pipeline costs vs. capacity and size

E. Decommissioning Cost
Offshore decommissioning is highly regulated by the Bureau of Safety and Environment Enforcement
(BSEE). According to BSEE, the process of decommissioning the well consists of safely plugging the
hole in the earths crust and disposing of the equipment used to support the production. BSEEs Idle
Iron policy keeps inactive facilities and structures from littering the Gulf of Mexico by requiring
companies to dismantle and responsibly dispose of infrastructure after they plug non-producing wells.
\\\
Platforms generally consist of two parts for decommissioning: the topside (the structure visible above
the waterline) and the substructure (the parts between the water surface and the seabed, or mudline).
In most cases the topsides that contain the operational components are taken to shore for recycling or
re-use. The substructure is generally severed 15 feet below the mudline, then removed and brought to
shore to sell as scrap for recycling or to be refurbished for installation at another location. An
alternative to onshore disposal is the conversion of a retired platform to permitted and permanently
submerged platform artificial reefs, commonly referred to as Rigs to Reefs (RTR). Based on BSEE
statistics, as of July 1, 2015, 470 platforms had been converted to permanent artificial reefs in the Gulf
of Mexico.
drilling However, all of these are fixed platforms located in shallow water.
To date, of all the GOM offshore platforms decommissioned only two were floating production units
located in water depths of 1,000 feet and deeper: ATP Innovator (semi) and Anadarko Red Hawk (spar).
ATP Innovator decommissioning involved disconnecting 10 riser-umbilical and 12 mooring lines, and
towing the Innovator to Ingleside, TX. The platform originally was built and converted from a Rowan
deep water semi drilling rig with an estimated cost of $300MM. IHS estimated the decommissioning
cost netted to scrape material is 45% of topside installation cost and 90% of semi hull installation. This is
equivalent to approximately $30MM.

127
EIA UPSTREAM COST STUDY

Anadarkos Red Hawk platform is the first cell spar deployed in the deep water GOM and made history
as the deepest floating production unit (FPU) ever decommissioned in the GOM. To reduce cost and
time spent hauling the structure from its location to onshore, Anadarko chose the Rigs to Reefs
program which previously had only been applied to shallow water fixed platforms. The original cost of
Red Hawk spar is estimated at $298MM and the conventional decommissioning cost is estimated at 45%
of topside installation cost and 50% of spar hull installation cost. By applying the Rigs to Reefs program
and sinking the hull to a nearby block, IHS estimates the decommissioning cost could be reduced by 28%
to $15MM from the conventional $21MM cost.

In general, IHS QUE$TOR estimates offshore deep water well decommissioning costs to be 10% of
installation cost. In other words, if installation is 90% of total D&C then decommissioning costs are 9%
of total well costs.

F. Operating cost
The deep water operating cost mostly involves floating production platform operating and maintenance.
Typically, a spar at 5,000 feet of water depth can have a monthly operating cost between $3MM to
$4MM. A semisubmersible is
more expensive to operate
compared to a spar or TLP.
Subsea tiebacks experience the
least operating expense, and
most of the cost incurred by
production handling agreement
(PHA) fee is paid to the host
platform. For floating
production platforms, the major
operating cost components are
platform inspection and
maintenance, operating
personnel, and insurance cost.
GOM operators are required to
purchase loss of production
Figure 9-49: Total lifecycle project LOE costs
insurance (LOPI) to cover the
production loss due to platform shut-ins and evacuations during hurricane season. Figure 9-49 provides
a total lease operating cost (LOE) cost comparison of the four selected offshore projects by development
concept.

G. Deepwater GOM cost trends


Because of the large scale of capital investment required to develop deep water fields, deep water GOM
\\\
operators are more pressured than US unconventional operators to increase efficiency and reduce cost.
We estimate that an approximate 20% capex cut is required to move unsanctioned projects in the GOM
Lower Tertiary play to a $60/bbl breakeven. With efficiency gains being rapidly realized in the US
unconventional plays, with operators focusing only on their first-tier prospect inventory and

128

drilling
EIA UPSTREAM COST STUDY

simultaneously delivering productivity improvements, the key question for the deep water GOM is how
quickly and to what degree operators can realize similar efficiencies.

IHS is projecting a 15% reduction in deep water costs in 2015, followed by a marginal average increase
of about 3% in overall deep water costs from 2016 to 2020 in nominal terms. Cost deflation is material
in many areas impacting deep water costs. This is particularly so in the rig market, where a rig overbuild
long forecast for 201516 is now colliding with reduced demand and resulting in quickly falling day
rates.

The three largest components of deep water capital costs are steel (~32% of deep water capital costs),
equipment (~21%), and rigs (~13%). Costs associated with all three components have declined into
2015, as the deep water market reacts to a weaker oil price environment and oversupply in many
segments.

Key drivers of cost reduction drilling rig

Going forward, contrary to the increasing rig supply result from overbuild during the last few years, rig
demand is falling. Operators are looking to reduce and delay expenditures to shore up portfolio returns
in response to a weaker oil price.

For the 3,001 feet to 7,500 feet


segment, IHS projects that fixed
rates are expected to continue
declining over 2015, becoming
essentially flat from 2016 to 2019,
and gradually recovering after
2017 (Figure 9-50).

While development drilling


proceeds on a robust queue of
sanctioned deep water projects,
reduction in exploration spend,
Figure 9-50: GOM deep water day rate forecast and therefore drilling, has more
limited near-term impact on
operator portfolios (making exploration easiest to cut first). Mid- to long-term implications can be quite
significant if deep water portfolios are not adequately restocked with new discoveries.

The most abrupt manifestation of the supply-demand disconnect in the rig market has been the early
termination of a number of rig contracts. With drilling rigs being a contracted service that cannot be
repurposed, cancellations will reduce exploration plans and add to the expectation that the re-
contracting of rigs with lower day rates can be achieved in an oversupply environment.
\\\
Key driver of cost reduction steel
Steel is the largest component part for deep water project costs. Steel prices have been declining for
several years as a result of oversupply. IHS suggests that steel prices are at or near their low point in

drilling 129
EIA UPSTREAM COST STUDY

Europe, Asia, and North America, with a tepid rally likely by the end of the year. Overall, this points to
the steel market being a buyers market for at least the next 18 months.
Specific to deep water project costs, steel costs directly impact deep water costs through a number of
required components that rely on steel:
notably facilities, topsides, offshore loading,
drilling, and subsea equipment. To assess this
broad impact, the IHS Capital Cost Services
carbon steel index tracks four specific product
groups: (1) line pipe, (2) structural steel, (3)
concrete reinforcing bar (rebar), and (4) oil
country tubular goods (OCTG), with OCTG
including both tubing and casing composed of
carbon steel or steel alloys. Based on this
index, we are modeling about a 16% cost
decrease in steel in 2015 versus 2014. Beyond
2015, a recovery in the steel market is
expected, with costs increasing approximately
11% in 2016 over 2015. More modest average
annual increases of about 3% are expected in
Figure 9-51: Steel cost forecast 201720. (Figure 9-51)

Key driver of cost reduction equipment


Included in oilfield equipment costs are turbines, exchangers, tanks and pressure vessels, pumps, and
compressors with restrictive standards and specifications for the oil industry. IHS is projecting declines
in upstream equipment costs over the next two years, followed by a moderate recovery between 2017
and 2020. As a result, in regards to deep water project modeling, we are forecasting an approximate
\\\
14% decrease in costs in 2015 compared to 2014 and a further 5% decrease in 2016. This is followed by
average annual increases in equipment costs of about 5% between 2017 and 2020.

The new deep water cost base


In addition to rigs, steel, and equipment, other key (but much smaller) components of deep water
project costs include engineering and project management (EPM), subsea facilities, installation vessels,
bulk materials, construction labor, freight, and yards and fabricationall of which are monitored in
detail by the IHS Capital Cost Service. In aggregate, and based on all these cost elements, we are
forecasting an approximate 15% decrease in non-equipment related capital costs in 2015, a 2% to 4%
dropdrilling
in 2016, and a modest recovery over the 201720 period.

Variations in cost indexes at a regional level are not insignificant. As a result, project level implications
associated with this cost decrease are not uniform and tend to vary by play. Nevertheless, in aggregate
within the global deep water, re-running economics for unsanctioned deep water projects with the new
lower cost structure does result in an average $5$10/Boe reduction in breakeven economics. This will

130
EIA UPSTREAM COST STUDY

not be considered an insignificant reduction as companies look to move to the next tranche of
developments past Final Investment Decision.

H. Key Take-Aways
Within the GOM deep water, substantial capital cost reductions are required in some plays to deliver
breakeven economics at $60/barrel, in addition to assumed reductions in operating cost. To achieve
$40/barrel breakeven costs, a more substantial additional capital expenditure cut is required. This may
be very difficult to achieve and many new discoveries may not be sanctioned. We estimate that an
approximate 20% capex cut is required to move unsanctioned projects in the US Gulf of Mexico Lower
Tertiary play to a $60/barrel breakeven, and at least a 30% cut to reach $40/barrel breakeven.

With efficiency gains being rapidly realized in the US unconventional space and with operators focusing
only on their first-tier prospect inventory and simultaneously delivering productivity improvements
(with one, of course, influencing the other), the key question for the deep water is how quickly and to
what degree can similar efficiencies be realized given the lack of critical mass and diversity of projects.

IHS Energy is forecasting an approximate 15% reduction in deep water costs in 2015, with an
approximate additional 3% reduction in 2016, and a modest recovery in nominal terms from 2017 to
2020.

131